More than any other condition, autism spectrum disorders (ASDs) are a group of diseases that affects families. They have profound effects on the individual with the disorder, and significantly alter the lives of parents, siblings, the family, and the community. The extended nature of ASDs has special significance to the geneticist, the medical specialist whose role is to identify causes of inherited conditions and provide counseling and guidance to the patient and family. To the geneticist, ASDs are a symptom complex in which impairments in socialization and communication occur in association with stereotypic behaviors. This symptom complex is a “final common pathway”, the result that multiple causes have on the brain. The goal of performing a genetic evaluation in an individual with ASD is: (1) to identify one of these underlying causes so that the family can learn more about the individual’s likely future functioning and potential medical problems, (2) to provide genetic counseling, which will inform the family about the likely recurrence risk of an ASD in future progeny, and (3) to provide an explanation of “why this happened.” For these reasons, evaluation by a geneticist should be offered as part of the work-up of all individuals diagnosed with ASD.
The fact that genetic factors clearly play an important role in the etiology of most cases of ASD is supported by multiple lines of evidence (Muhle R, Trentacoste SV, Rapin I. The genetics of autism. Pediatrics 2004;113:e472–e486). For instance, according to the Centers for Disease Control, in 2007, the prevalence of autism at age 8 in 6 areas of the United States was approximately 1 in 150 (MMWR, February 9, 2007 / 56SS01; 1-11). However, after the birth of a child with an ASD, the empiric recurrence risk (based on observation of thousands of families) for full siblings is between 4 and 7%, much higher than would be expected in the general population. Further, if a second child has autism, the recurrence risk rises to 25-35%. These increases point to the role genetic factors play in the etiology of ASD.
The geneticist does not make the diagnosis of an ASD. Rather, this specialist is part of the interdisciplinary team of health care professionals who are involved in the diagnosis and management of such individuals. Ideally, a referral is made to the geneticist after the diagnosis has been confirmed. The geneticist’s evaluation should occur while other evaluations are being performed and a therapeutic plan is being developed.
The Genetic Evaluation
What can a family expect when their child is referred for a genetic evaluation? What does the clinical geneticist actually do? In most cases, the genetic evaluation begins with a complete history, focusing on issues that might have contributed to the affected individual’s (or proband’s) condition. Although, as previously stated, genetic factors play a role in most cases of ASD, in some cases, environmental exposures occurring prior to birth have been implicated. Specifically, the geneticist is concerned about exposure to intrauterine infections such as Rubella (German measles) or cytomegalovirus, prenatal exposure to certain drugs and chemicals, such as the anti-seizure medication Valproic Acid and alcohol. Information about the proband’s general health, age at onset of symptoms, presence or absence of language and developmental regression and of seizures, and the age at which the diagnosis was made is also obtained.
Next, a complete family history is taken. Assembled as a pedigree, a pictorial representation of the family history, information about at least three generations is recorded. The family history includes details about the presence of ASDs, as well as other conditions causing developmental and behavioral disabilities; also information about the diagnosis of any genetic disorder, number of miscarriages and of childhood deaths in these related individuals is also obtained.
This is followed by a complete physical exam. When examining the proband, the geneticist searches for subtle clues that suggest the presence of a genetic disorder known to be associated with ASD (a partial list appears in Table 1). For instance, does the proband have café-au-lait macules, pigmented spots on the skin known to be associated with neurofibromatosis, or hypopigmented spots in the shape of ash leaves, which suggest tuberous sclerosis (the search for ash leaf spots is aided by the use of a Woods lamp)? Are there unusual facial features such as a prominent forehead or jaw, or large ears that point to fragile X syndrome? Is the head circumference larger (suggesting Sotos syndrome or the PTEN-associated disorders) or smaller (supporting the effects of a teratogen) than would be expected for a child of that age? Are the joints lax or, in boys, are the testes large (both features of fragile X)? Does the proband have unusual hand movements such as flapping (associated with fragile X), ataxic, puppet-like movements (seen in Angelman syndrome) or “hand washing” movements (noted in girls with Rett syndrome)?
The Genetic Work-Up
Following completion of the history and physical exam, an assessment is made about the likely cause of the proband’s condition. Specifically, in order to direct the laboratory evaluation and counseling, the geneticist must decide whether the proband has a primary ASD (an ASD not associated with an underlying genetic disorder or environmental teratogen), or an ASD that is secondary to a prenatal exposure (such as congenital rubella syndrome or fetal alcohol spectrum disorder) or to the presence of a genetic disorder, such as neurofibromatosis, fragile X or Rett syndrome. Between 80 and 90% will be judged to have a primary ASD, while 10-20% will have a secondary ASD.
Based on the outcome of the assessment, the geneticist will decide the laboratory tests, if any, that should be performed. As noted in Table 2 (and described in Schaefer GB, Mendelsohn NJ. Genetics evaluation for the etiologic diagnosis of ASDs. Genet Med 2008;10:4–12.), the testing should be done in stages or tiers, in which tests are ordered in sequence, based on results of previous evaluations. If an ASD is due to an underlying genetic disorder, direct DNA testing should be performed (or, in the case of neurofibromatosis, the diagnosis should be confirmed using the scoring system described by Riccardi [Ferner RE. Neurofibromatosis 1 and neurofibromatosis 2: a twenty first century perspective. Lancet Neurol. 2007;6:340-351.]). If a congenital infection is suspected, serologic testing or polymerase chain reaction may be helpful in confirming the etiology, while the diagnosis of a teratogenic syndrome, such as fetal alcohol spectrum disorder or valproic acid embryopathy, is confirmed on the basis of the history of exposure and the presence of characteristic features.
As already noted, the majority of probands will have primary ASD, with no features suggestive of an accompanying genetic disorder. It is this population that presents the greatest challenge to the geneticist. In these cases, discovering information that will prove helpful to the family is often difficult and elusive. However, new tools, such as microarray comparative genomic hybridization, have offered hope that the cause will be identified in more individuals with a primary ASD
“Silent” inborn errors of metabolism have been postulated to be a cause of primary ASDs. In 1994, Laszlo et al. reported that 43% of children who met the DSM III criteria for ASD had elevated levels of lactic acid, a non-specific biochemical marker of underlying abnormalities in glucose metabolism (Laszlo A, Horvath E, et al.: Serum serotonin, lactate and pyruvate levels in infantile autistic children. Clinica Chimicta Acta 1994; 229:205-207). Although subsequent studies have identified a smaller percentage of probands with elevated lactate levels, routine testing for lactate and pyruvate should be part of first tier testing. If levels are normal, no further evaluation is indication; if abnormal, a series of tier 2 tests should be performed.
The need for analysis of chromosomes has been well established in the evaluation of the cause of an ASD. Numerous small (submicroscopic) deletions and duplications, called copy number variants (CNVs), have been associated, including duplication of a portion of chromosome 15 (15q11.2), seen in 3% of cases, and a deletion in the short arm of chromosome 16 (16p11.2), seen in 1%. In the past, the approach to identify these errors has always included high resolution chromosome analysis accompanied by fluorescent in situ hybridization (FISH) looking for errors at the end of the chromosomes (subtelomeres). More recently, microarray comparative genomic hybridization, a powerful technique 4 to 5 times more sensitive at identifying CNVs, has begun to supplant these previously used techniques. Recent studies have indicated that using array CGH, CNVs can be identified in 10% of individuals with sporadic autism beyond what would be identified by standard chromosomal testing (Bejjani BA, Shaffer LG: Clinical Utility of Contemporary Molecular Cytogenetics. Annual Review of Genomics and Human Genetics 2008, 9:71-86).
Although array CGH has the potential to identify CNVs, the technique cannot identify mutations within genes. As such, even in the absence of clinical features that suggest the diagnosis, because of the high yield in this population and the implications to other members of the family, all probands should have DNA testing for fragile X, and all females should be tested for MECP2 associated disorder (Rett syndrome).
According to Schaefer and Mendelsohn, the aggregate result of tier 1 testing is that between 22 and 33% of individuals with primary ASD would have an identifiable cause of their condition. This includes between 10 and 15% with CNVs, 5% with fragile X, 2 to 3% with MECP2-related disorders, and 5 to 10% with other causes.
Based on results received, following this first tier, a second tier of testing should be performed. As noted in Table 2, if the results of metabolic testing are abnormal, a more complete evaluation is needed to pinpoint the specific error in metabolism. If a CNV or genetic mutation is identified in the proband, the parents should be tested to see if this is inherited or occurred de novo (spontaneously).
Finally, depending on the results of the tier 2 testing, a group of tests in the third tier might be performed. This includes the performance of an MRI to identify an underlying structural brain anomaly, more complete metabolic testing, and, as indicated, DNA testing of additional members of the proband’s family.
Following completion of this work-up, the family is invited back for genetic counseling. If a specific etiology, such as fragile X, has been identified, counseling is provided for that condition. If the proband has primary autism and work-up has revealed a genetic cause, such as a CNV that is not present in either parent, a low recurrence risk, on the order of 1%, can be cited. If the work-up has failed to reveal any genetic cause, an empiric recurrence risk for full siblings of 4% if the proband is female, and 7% if the proband is male, can be given. If it has been found that two children in the family are affected, the empiric recurrence risk rises to between 25 to 35%. This visit is also an opportunity for the family to ask questions of the geneticist.
Follow-up with the geneticist is necessary, especially when no underlying etiology has been identified. Because technology in the field of clinical genetics is advancing so rapidly, it is possible that new tests will become available that will identify an etiology in children in whom currently no etiology is identifiable. For this reason, periodic reevaluation on an annual or biannual basis is important.