NEW YORK (GenomeWeb News) – In a study appearing online last night in PLoS ONE, researchers from Boston Children's Hospital, Harvard Medical School, and elsewhere reported on a gene expression signature in the blood that appears to hold promise for diagnosing autism spectrum disorder.
The team, led by Isaac Kohane and Louis Kunkel, used microarrays to profile gene expression patterns in blood samples from 66 boys with ASD and 33 age- and sex-matched controls, uncovering hundreds of differentially expressed genes in the blood of children with ASD. A subset of these — a 55-gene expression signature that the study's authors dubbed ASD55 — accurately distinguished between the male cases and controls just over three-quarters of the time in the original training set.
When they tested expression of the ASD55 genes in blood samples from another 104 boys and girls with ASD and 82 without, investigators found that blood expression profiling showed similar accuracy in male children, correctly classifying 73 percent of the boys as cases or controls. The accuracy was lower for girls (64 percent), but came in at 68 percent for the group as a whole.
"By looking at this 55-gene signature, which can capture disruptions in multiple pathways at once, we can say with about 70 percent accuracy, 'this child does not have autism,' or 'this child could be at risk,'" Kohane said in a statement.
If similar gene expression profiles are present before ASD symptoms arise, he and his colleagues explained, the blood-based gene expression profiling approach could prove a fast and affordable way to identify those at risk of the disease and perhaps provide some with earlier treatment.
"Our result suggests that the use of blood expression profiling for ASD detection may be feasible," they wrote. "Further study is required to determine the age at which such a test should be deployed," they added, "and what genetic characteristics of ASD can be identified."
As GenomeWeb Daily News sister publication BioArray News reported last year, the Massachusetts-based diagnostic start-up company SynapDx holds an exclusive license on Boston Children's Hospital's expression-based strategy for ASD testing.
A host of past studies support the notion that ASD is a consequence of both genetic and environmental factors. But a broad range of mutations and alterations have been implicated in the neurodevelopmental condition, making it difficult to come up with a broadly applicable genetic diagnostic test.
For its part, the Boston Children's Hospital-led team reasoned that they might have more luck classifying cases and controls based on RNA profiles, since gene expression, like ASD itself, is thought to be influenced by both genetics and the environment.
In developing the gene expression classifier, researchers used Affymetrix arrays to profile gene expression patterns in blood samples from the initial cohort — comprised of 66 male children with ASD between the ages of about five and 10 years old and 33 unaffected males in more or less the same age range. They also profiled expression patterns in 104 male and female cases and 82 controls from the validation cohort.
In samples from the male children in the original cohort, the team tracked down 489 differentially expressed genes. Chief among them were genes in immune, inflammation, and signaling pathways. The latter group included some neurotrophin signaling genes, which are believed to contribute to neural development learning and memory.
From there, researchers whittled down their list of candidates to come up with the 55 most promising potential ASD predictors. Blood expression profiles for these ASD55 signature genes correctly classified cases and controls nearly 68 percent of the time in the validation group, they reported.
Still, the approach picked up more authentic ASD cases in boys than in girls. For males in the validation group, the test showed 90 percent sensitivity and 43.8 percent specificity (72.7 percent accuracy) compared with 50 percent sensitivity and 73.5 percent specificity (63.8 percent accuracy) in girls.
In contrast, the study's authors noted that microarray testing for ASD-associated chromosomal alterations typically picks up around 7 to 10 percent of cases, prompting them to argue that the ASD55 predictor "compares favorably to the DNA-based tests currently proposed for ASD diagnosis."
"[W]e expect that larger studies can be used to determine whether particular characteristics of ASD can be classified or predicted from a gene expression signature (e.g. seizures and language delay)," they concluded, "and thereby improve individualized treatment in the near future."