SEARCH

SEARCH BY CITATION

The debate over autism testing continues

How much testing is appropriate after a child is diagnosed with autism, a behaviorally defined condition? In recent years, this question has generated much debate—and the conversation continues as whole genome testing technologies advance. The controversy shows no sign of abating, if discussion on autism spectrum disorder (ASD) testing at the March American College of Medical Genetics (ACMG) meeting is any indication.

Thumbnail image of

Should children with autism receive genetic testing for the condition?

During interviews after the meeting, participants in a panel discussion on spotting clues that point to inborn errors of metabolism or biochemical defects in patients with either ASD or intellectual disability brought up points of debate. These include high cost, relatively low yield, and the sometimes-remote possibility of finding an effective treatment for an underlying cause. Who actually benefits from testing—patients, parents, or researchers—is another consideration.

In an attempt to address some of these questions, the 2008 ACMG guidelines [Schaefer et al.] on autism aim for middle ground by establishing 3 tiers of genetic and other tests that identify a cause in an estimated 40% of patients with ASD. Tier 1 includes tests that authors say should be performed in all children with no obvious causes of autism. These may include chromosome tests, DNA studies for fragile X syndrome, and testing for certain metabolic conditions with indicative symptoms. No further testing is needed if these tests yield a cause. Tier 2 includes comparative genomic hybridization, methyl-CPG-binding protein 2 (MECP2) gene testing in girls, and phosphatase and tensin homolog (PTEN) gene testing in children with a large head circumference. Brain magnetic resonance imaging (MRI), chromosome 15 methylation, and serum and urine uric acid tests are part of Tier 3.

A Continuum of Opinion

  1. Top of page
  2. A Continuum of Opinion
  3. Another View
  4. An Argument for the ACMG Guidelines
  5. The Lure of Whole Genome Testing
  6. References

The ACMG panelists have varying views about autism testing in general and these guidelines specifically. Opinions fall along a continuum that ranges from discouraging genetic testing without some sign of a condition associated with ASD to encouraging metabolic testing for all children with autism.

Isabelle Rapin, MD, Professor of Neurology and a pediatric neurologist at Albert Einstein College of Medicine in the Bronx, New York, is on one end of the spectrum. She takes issue with genetic and metabolic tests that are not prompted by a clinical clue, she maintains. That is because she finds current tests are very rarely informative to treatment. “So the question is,” she says, “how many uninformative tests are needed to identify the rare unsuspected positive test that would help the child's management?”

In an essay distributed during the ACMG session, she urged clinicians to consider whether assays and studies serve patients directly or benefit parents and medical research. “Genetic testing is essential for research, rarely so for the individual, and a little more often but still rarely, for genetic counseling,” she adds. While she supports genetic testing for studies paid for by research funds, she comments, “some of the money ought to be spent testing interventions.”

Another View

  1. Top of page
  2. A Continuum of Opinion
  3. Another View
  4. An Argument for the ACMG Guidelines
  5. The Lure of Whole Genome Testing
  6. References

Metabolic geneticist Carol Greene, MD, Professor of Pediatrics and Director of the pediatric genetics clinic at University of Maryland Medical Center in Baltimore, is on the other end of the spectrum. She agrees with the concept of tiered testing embraced by the ACMG guidelines, but favors expanding the first tier to promote certain basic metabolic tests for all autistic children. That is because these tests are inexpensive relative to genetic assays, and can make a large difference in the lives of children who do have identified, treatable metabolic conditions and for their parents, who face a high risk for recurrence in siblings.

The metabolic tests include blood count, liver function, electrolytes, and plasma amino acids, plus tests for large amounts of the mucopolysaccharide heparan sulfate, and creatine phosphokinase for Duchenne muscular dystrophy. These tests yield diagnoses associated with autism in just 1 to 3 percent of cases, explains Dr. Greene, but she adds, “you can't find what you don't look for.” Among the diagnoses are phenylketonuria and homocystinuria. Both are treatable, have no specific signs and symptoms, and have a 25% risk of recurrence in siblings, she points out.

Biochemical geneticists who were also ACMG panelists note that metabolic testing is at the core of their work. “As metabolic physicians, we feel sometimes obligated to do many tests,” explains panel co-chair Nicola Longo, MD, PhD, Professor and Chief of Medical Genetics at the University of Utah in Salt Lake City.

An Argument for the ACMG Guidelines

  1. Top of page
  2. A Continuum of Opinion
  3. Another View
  4. An Argument for the ACMG Guidelines
  5. The Lure of Whole Genome Testing
  6. References

ACMG guidelines author G. Bradley Schaefer, MD, Professor of Genetics and Pediatrics at the University of Arkansas for Medical Sciences in Little Rock, says his approach is guided by recognition that autism testing requires a lot of time and money. Noting he was a “vocal audience member” during the meeting, he responded to Dr. Rapin's view that genetic testing borders on research.

Dr. Schaefer agrees with Dr. Rapin that genetic testing really serves families, not individual patients—but that is a valuable service for both families and society as a whole, he argues. Knowing the causes of ASD can be empowering for families. “Families seek out geneticists because they want to know why their child has an ASD,” he maintains. “It's not research to find out why a kid has a condition.” And while widespread but unfounded fears that the MMR vaccine causes autism persist, testing can allay this dangerous misperception in society at large.

The most common reason for testing is family planning, he notes. In families with one child with ASD, the risk of having a second child with ASD is 4% to 8%. If two children have autism, the risk of recurrence in a third child skyrockets to 30% to 50%. Families want to know if the chance of another ASD child is low or akin to a coin toss, notes Dr. Schaefer. Even if testing does not influence care for autism, it can uncover medical comorbidities and allow planning for the future.

The debate about testing is mostly about cost, he maintains. On average, autism testing costs about $5,000, and with all 3 ACMG-recommended tiers, that figure rises to roughly $15,000. One of the goals of the tiered system advocated by the guidelines is to “give the biggest bang for the buck,” he notes, adding, “If all the tests I suggest could be done for $100 and a kid just spits in a cup, we wouldn't be having this debate.” In the end, parents make testing choices. They must understand the demands and limits of testing, notes Dr. Schaefer.

If all the tests I suggest could be done for $100 and a kid just spits in a cup, we wouldn't be having this debate.—G. Bradley Schaefer, MD

The Lure of Whole Genome Testing

  1. Top of page
  2. A Continuum of Opinion
  3. Another View
  4. An Argument for the ACMG Guidelines
  5. The Lure of Whole Genome Testing
  6. References

Usually geneticists treat children of families who are searching desperately for answers about the cause of autism. “We do so much testing because helping families is the right thing to do. But from a an analytical point of view, it's not logical because the yield of testing for ASD is so low,” explains panel co-chair Gerard T. Berry, MD, director of the metabolism program at Children's Hospital Boston. As a biochemical geneticist, he performs extensive autism testing in his practice.

Dr. Berry calls for a new paradigm in autism testing fueled by the promise of whole genome sequencing, which is currently a research endeavor. For now, geneticists have comparative genome hybridization (CGH), also called chromosomal microarray. It may have 3 times the detection rate for chromosomal genetic changes related to ASDs than standard G-banded karyotype, according to a recent study from Children's Hospital Boston and Boston-based Autism Consortium researchers [Shen et al., 2010]. In their report on the research involving chromosomal microarray analysis of 933 patients with ASD, the researchers estimate that standard testing misses identifiable chromosomal abnormalities in at least 5% of ASD cases.

Current ACMG guidelines place CGH to the second tier of testing, but Dr. Schaefer expects that recent research findings will elevate CGH to the first tier in a future version.

Dr. Greene sees pros and cons to CGH, which costs about $2,000. If a child has no specific clues indicating genetic cause for autism, CGH is cheaper and easier than the current testing paradigm. But if a child's ASD results from an abnormal number of chromosomes, as with Klinefelter syndrome or phenylketonuria, a cheaper test could identify the cause. “If that's the case, a family might regret opting for the more expensive test,” she points out.

While a less expensive and more effective testing strategy like CGH is always good news, whole genome sequencing's potential to end controversy surrounding autism testing could become yet another topic of debate, Dr. Greene predicts.

References

  1. Top of page
  2. A Continuum of Opinion
  3. Another View
  4. An Argument for the ACMG Guidelines
  5. The Lure of Whole Genome Testing
  6. References
  • Schaefer GB, Mendelsohn NJ; Professional Practice and Guidelines Committee. 2008. Clinical genetics evaluation in identifying the etiology of autism spectrum disorders. Genet Med 10: 301305.
  • Shen Y, Dies KA, Holm IA, Bridgemohan C, Sobeih MM, Caronna EB, Miller KJ, Frazier JA, Silverstein I, Picker J, Weissman L, Raffalli P, Jeste S, Demmer LA, Peters HK, Brewster SJ, Kowalczyk SJ, Rosen-Sheidley B, McGowan C, Duda AW 3rd, Lincoln SA, Lowe KR, Schonwald A, Robbins M, Hisama F, Wolff R, Becker R, Nasir R, Urion DK, Milunsky JM, Rappaport L, Gusella JF, Walsh CA, Wu BL, Miller DT; Autism Consortium Clinical Genetics/DNA Diagnostics Collaboration. 2010. Clinical genetic testing for patients with autism spectrum disorders. Pediatrics 125: e727e735.