**6. Status**

Potential endophenotypes related with language and social communication

A continuous distribution of social deficits has been described in the general population, being more frequent in boys than in girls and more prevalent in brothers and fathers than in sisters and mothers of patients with ASD [54]. The estimated frequency of such disorders in siblings ranges from 2.6% to 61.7% [55-59].

It has been suggested that siblings later diagnosed with ASD may be distinguished from other siblings and low-risk infants by unusual eye contact, visual tracking, disengagement of visual attention, orienting to name, imitation, and delayed social expressive and receptive language [60].

**i.** Be identified for genetically related disorders [48-51].

phenotypes that may fulfill the requirements of endophenotypes.

proposed for neuropsychiatric disorders.

356 Autism Spectrum Disorder - Recent Advances

**4. Research course**

anatomy, or neurophysiology.

ranges from 2.6% to 61.7% [55-59].

**5. Methods**

**6. Status**

Thus far, the identification of endophenotypes has been successful in several fields of Medi‐ cine. Endophenotypes were useful in the identification of genes for QT syndrome, idiopatic hemochromatosis, juvenile myoclonic epilepsy, and familial adenomatous intestinal polypo‐ sis. Biochemical assays, physiological measures and challenges contribute as primary evidence of pathology in the detection of increased risk to diabetes, hypercholesterolemia, obesity, hypertension, and osteoporosis [48, 52, 53]. Furthermore, different endophenotypes have been

First degree relatives of patients are key elements in the study of endophenotypes. They are considered at-risk individuals. Even so, only a relatively small percentage of siblings of an individual with ASD will develop such disorders, but a larger proportion is expected to exhibit

It has been shown that siblings of patients with ASD as well as other first degree relatives sometimes exhibit social, communication, and/or learning deficits, and ASD are more frequent in them than in the general population. Several possible endophentoypes for ASD have been proposed so far and are related with language and communication and/or with cognition,

Medline database was searched for articles related with autism and autism spectrum disorder endophenotypes (query terms: autism spectrum disorders and endophenotypes, neuroanat‐ omy, neurophysiology, endocrinology, cognition, social communication, and language. OMIM compendium was searched for disease and gene names. We will discuss potential endophenotypes based on studies in which one or more of their characteristics were fulfilled.

A continuous distribution of social deficits has been described in the general population, being more frequent in boys than in girls and more prevalent in brothers and fathers than in sisters and mothers of patients with ASD [54]. The estimated frequency of such disorders in siblings

It has been suggested that siblings later diagnosed with ASD may be distinguished from other siblings and low-risk infants by unusual eye contact, visual tracking, disengagement of visual

Potential endophenotypes related with language and social communication

Moreover, there are reports of poor social and communicative performance in parents without a diagnosis of ASD and siblings of children with such disorders [61]. Families with multiple affected individuals were compared with single-incidence autism families, and Down syndrome families as controls. In most cases with multiple affected relatives, both parents had autism-related characteristics. This was not observed in single-incidence families [62].

Furthermore, children with ASD and their parents have been shown to have an impaired performance in rapid automatized naming. Siblings of patients had intermediate performances when compared with controls [62, 63].

Poor verbal imitation and reduced response to vocal approaches have also been described in children with ASD or children who later developed ASD [64, 65]. In a study, 100% of the 12 month-old infants with no risk for ASD responded on the first or second name call, but 14% of at risk infants responded later. At two years of age developmental problems were identified in many of the children who failed to respond [66].

Impairment in joint attention abilities at age 20 months was also associated with later social and language symptoms [67]. At that same age, children with ASD failed to orient towards biological motion [68]. In addition, affected children who were three to four years old per‐ formed worse not only in social orienting and joint attention, but also in attention to the distress of someone else when compared with children without ASD [69].

There are genes that have been associated with language and social impairment in children with ASD and their siblings. This may be illustrated by a familial mutation of the MET gene. The mutation consists of a deletion of 4 exons that cause a frameshift and a premature stop codon. On the homologous chromosome 7 the patient carried an allele that was associated with a decrease in MET expression.

A sibling who was also a carrier of the deletion had language and social impairment [70]. MECP2 gene has been associated with difficulty in recognizing emotional expressions and less time spent looking at facial features [71]; it has also been described as a regulator of the expression of genes involved in social behavior [72]. In animal models, social deficits are observed when the activity of specific genes is lacking, such as FMR1 and Mu-opiod recepetor gene (OPRM1) [73].

#### **Cognitive, anatomical and neurophysiological candidate endophenotypes**

Atypical patterns of brain activation, for example, in the fusiform gyrus and the amygdala have been identified in individuals with ASD during face processing [74, 75]. In adults, parents and siblings of individuals with ASD were less able to discriminate subtle differences between faces than control individuals, but performed better than adults with ASD. Relatives were significantly worse at identifying expressions of fear and disgust than controls and did not show the typical sensitivity to direct eye gaze direction as opposed to averted eye-gaze, a characteristic observed in adults with ASD [76]. In a prospective study, 6 to 10 month-old infants were asked to view faces with eye gaze directed toward or away from the infant. Characteristic components of event-related potentials were associated with ASD diagnosed later, at 36 months [77]. In one study, functional magnetic resonance imaging and eye tracking in unaffected siblings of individuals with ASD were performed. There were important differences in gaze fixation and brain activity in response to images of human faces when unaffected siblings were compared with typically developing controls. Siblings performed in a similar way than the ASD patients. When amygdala volume was compared, individuals with ASD and their siblings had similar volumes, which were significantly reduced when compared with volumes in the control group [78]. Small differences were independently found in the left amygdala when subjects with ASD or siblings were compared with controls [79]. Moreover, boys with ASD and unaffected brothers of individuals with ASD showed deficits in a visual attention task. Atypical fronto-cerebellar activation was identified in patients and unaffected brothers when they were compared with controls [80, 81].

Differences in gray matter in the cerebellum, parietal lobe, and left occipital lobe have been described between patients with ASD and controls. In some cases, differences between siblings without ASD and controls have also been reported [79, 82]. Furthermore, deficits in frontal, temporal, and occipital lobes have been identified in discordant monozygotic twins. Eight out of nine of the unaffected twins showed a social or language delay [83]. The functional magnetic resonance imaging response to happy facial expression was different between unaffected siblings and healthy controls without a family history of ASD. Here there were no statistically differences in response between unaffected siblings and patients with ASD [84].

The CNTNAP-2 gene (at 7q35 region) is mutated in individuals with intellectual disability, seizures, autistic features, and language impairment. Connectivity in the frontal lobe has been associated with CNTNAP-2 genetic variants [85].

White matter structure was analyzed in a voxel-based diffusion tensor imaging study. Children with ASD and their unaffected siblings had significantly reduced white matter fractional anisotropy values in the frontal parietal and temporal lobes when compared with controls. There were no significant differences in white matter structure between the ASD and sibling groups [86].

In other studies, children with ASD and their parents showed eye movement abnormalities that suggested a poorer spatial accuracy with respect to controls. The authors proposed that the spatial working memory may be affected in parents [87].

It has been suggested that ASD is characterized by a tendency towards local rather than global information processing. This has been called weak central coherence, which has also been reported in fathers of boys with ASD, as indicated by piecemeal processing in four tests [88]. In addition, some parents preferred nonsocial activities and ability in detail-focused process‐ ing, as is expected in subjects with ASD. A questionnaire completed by parents discriminated between ASD individuals and controls, but did not differentiate siblings [89]. In addition, atypical activation of temporal and frontal regions during the Embedded Figures Task was found in subjects with ASD and in unaffected siblings. The reduced activation on the ASD group when compared with controls was correlated with the severity of reciprocal social interaction deficits [90].

infants were asked to view faces with eye gaze directed toward or away from the infant. Characteristic components of event-related potentials were associated with ASD diagnosed later, at 36 months [77]. In one study, functional magnetic resonance imaging and eye tracking in unaffected siblings of individuals with ASD were performed. There were important differences in gaze fixation and brain activity in response to images of human faces when unaffected siblings were compared with typically developing controls. Siblings performed in a similar way than the ASD patients. When amygdala volume was compared, individuals with ASD and their siblings had similar volumes, which were significantly reduced when compared with volumes in the control group [78]. Small differences were independently found in the left amygdala when subjects with ASD or siblings were compared with controls [79]. Moreover, boys with ASD and unaffected brothers of individuals with ASD showed deficits in a visual attention task. Atypical fronto-cerebellar activation was identified in patients and unaffected

Differences in gray matter in the cerebellum, parietal lobe, and left occipital lobe have been described between patients with ASD and controls. In some cases, differences between siblings without ASD and controls have also been reported [79, 82]. Furthermore, deficits in frontal, temporal, and occipital lobes have been identified in discordant monozygotic twins. Eight out of nine of the unaffected twins showed a social or language delay [83]. The functional magnetic resonance imaging response to happy facial expression was different between unaffected siblings and healthy controls without a family history of ASD. Here there were no statistically

The CNTNAP-2 gene (at 7q35 region) is mutated in individuals with intellectual disability, seizures, autistic features, and language impairment. Connectivity in the frontal lobe has been

White matter structure was analyzed in a voxel-based diffusion tensor imaging study. Children with ASD and their unaffected siblings had significantly reduced white matter fractional anisotropy values in the frontal parietal and temporal lobes when compared with controls. There were no significant differences in white matter structure between the ASD and

In other studies, children with ASD and their parents showed eye movement abnormalities that suggested a poorer spatial accuracy with respect to controls. The authors proposed that

It has been suggested that ASD is characterized by a tendency towards local rather than global information processing. This has been called weak central coherence, which has also been reported in fathers of boys with ASD, as indicated by piecemeal processing in four tests [88]. In addition, some parents preferred nonsocial activities and ability in detail-focused process‐ ing, as is expected in subjects with ASD. A questionnaire completed by parents discriminated between ASD individuals and controls, but did not differentiate siblings [89]. In addition, atypical activation of temporal and frontal regions during the Embedded Figures Task was found in subjects with ASD and in unaffected siblings. The reduced activation on the ASD

differences in response between unaffected siblings and patients with ASD [84].

brothers when they were compared with controls [80, 81].

358 Autism Spectrum Disorder - Recent Advances

associated with CNTNAP-2 genetic variants [85].

the spatial working memory may be affected in parents [87].

sibling groups [86].

Regarding body measurements, although there is a high degree of variability in head circum‐ ference of children with ASD, the rate of macrocephaly is increased in these patients, if compared with the general population. Mean standardized head circumference and frequency of macrocephaly were similar in patients with ASD and their parents. Increased head circum‐ ference was associated with ASD severity and with a late onset of language [91]. Macrocephaly appearing in the first year of life has been described in subgroups of patients with ASD, mental retardation, and/or language delay who carry mutations on the Phosphatase and tensin homolog (PTEN) gene. Numerous relatives of individuals with a PTEN mutation also had macrocephaly and mental retardation [92-94]. Mutations in the hepatocyte cell adhesion molecule gene (HEPACAM) have also been associated with macrocephaly and ASD [95].

Other researchers have documented impairment in cortical sensory processing in ASD cases. For example, it has been suggested that early auditory responses as well as transient memory are impaired in some children with ASD; auditory stimuli produce messages that are sent from the brainstem to the thalamus and into the auditory cortex. Event related potentials were found to be delayed in affected children, and indicated a deficiency to achieve typical maturational development of the auditory system in individuals with ASD [96, 97]. In addition, children who were at risk for ASD showed atypical neural lateralization to speech in their first year of life [98].

Impairment in cortical sensory processing has been found in patients with Rett syndrome. In addition, auditory and visual evoked potentials in mice that were heterozygous for a loss of function variant of the MeCP2 gene demonstrated impairments in sensory processing. For example, increased N1 amplitude, latency of auditory P2 component, and suppression of betaband activity were identified. There was a decreased suppression of induced gamma-band activity [99]. Gamma-band activity differences have been identified between parents of children with ASD and control adults [100].

Another area of potential endophenotypes is related with movement. There have been reports that suggest that specific motor developmental clues may be identified early in infants who will develop ASD. These signs tend to be persistent, and may be associated with abnormalities at the cerebellar, vestibular, cortical or fronto-striatal level [101]. Moreover, infants who were 18-month old and who had an older sibling with ASD were at higher risk for a delay in postural change and stability, language production and comprehension if compared with infants with no risk for ASD [56]. An association of motor anomalies and specific genes has been proposed. MeCP2, SHANK3, 22q13.3, Neurexin 1 (NRXN1) deletion, 15q11.1-q13.3, and KIAA0442 gene (known as Autism Susceptibility Candidate 2 or AUTS2) have been associated with hypotonia, stereotypes and/or motor delay; RNA-binding protein FOX1 (RBFOX1) has been associated with motor assymetry [102].
