Skip to main content
Premium Trial:

Request an Annual Quote

Gene Expression Profile Predicts Response to Risperidone in Autistic Children


By Molika Ashford

The expression of five genes is associated with response to the atypical antipsychotic risperidone in children with autism spectrum disorders, a new study by researchers at the University of California, Davis, has found.

The group, led by the university's MIND Institute, measured gene expression in the blood of 42 ASD subjects prior to their treatment with risperidone and compared these expression levels to subjects' response to treatment after eight weeks. A signature of five genes — GBP6, RABL5, RNF213, NFKBID, and RNF40 — was associated with improvement in behavioral issues post-risperidone treatment, as measured by the Aberrant Behavior Checklist-Irritability score system. The study was published online last month in The Pharmacogenomics Journal.

The authors call their study the "first to suggest that gene expression in blood is associated with and may predict the behavioral response to risperidone use in ASD." While it is their eventual goal to develop a test that would help select patients most likely to respond to risperidone, the researchers cautioned that much validation work is necessary before moving forward with diagnostic plans.

Risperidone was developed by Janssen and marketed under the brand name Risperdal. In 2006, risperidone was approved by the US Food and Drug Administration for the treatment of irritability in autistic children and adolescents. Since 2008, various drug developers have been selling generic risperidone.

While risperidone and other antipsychotics have limited effect on the core symptoms of autism spectrum disorders, they are effective in treating behavioral disturbances including anxiety, aggression, and sleep disturbance, Lisa Lit, a MIND Institute PhD and first author of the paper, told PGx Reporter.

Lit said that while risperidone is a great treatment for these issues, it also has side effects including increased lipids and diabetes. The researchers hoped to establish a gene expression signature to identify which children would not benefit from the drug, and could therefore avoid taking it.

"You don't want to be putting children on medications if they're not going to be effective. If we can identify ahead of time what's going to work versus what's not going to work, we have a little advantage when we're talking about treatment for these kids," she said.

The researchers recruited 42 children diagnosed with ASD, who had an initial ABC-I score above 18, which is close to the mean. Subjects could not have used antipsychotic medication within eight weeks of their entry into the study, and they agreed not to start any new drug or other treatment during the course.

Using Affymetrix GeneChip Human Exon 1.0 ST arrays, the group measured gene expression from peripheral blood samples before beginning risperidone treatment.

Treatment was started for all subjects at 0.5 mg at bedtime, with increasing dosage after 4 days if the drug was tolerated and symptoms supported a greater dose up to 1.5 mg.

The team measured symptoms using the ABC-I scoring system after eight weeks of treatment and used the difference between initial score and final score to define drug response. According to Lit, eight weeks is considered the point where it is evident whether risperidone is working or not. Many of the subjects showed a pretty strong response by that point, she said.

To establish what, if any, gene expression patterns were associated with response, the researchers separated out the 17 patients with the most extreme responses according to their ABC-I scoring differential, grouped into high responders and low responders. Then they moved out to establish a pattern that could stratify the range of responses in the full patient group, the authors wrote.

For the extremes, 89 exons were initially identified as differentially expressed in low and high responders. Of these, five were significantly correlated with ABC-I score change across all 42 subjects, GBP6, RABL5, RNF213, NFKBID, and RNF40.

Pathway analysis of these markers did not yield any significant findings, the authors wrote.

There was no overlap between the genes found by the MIND researchers and those from two other studies published in 2010 that looked at genetic associations with risperidone response in ASD and schizophrenia. The authors wrote that this disparity may reflect "convergent downstream biological mechanisms across multiple genetic backgrounds" that contribute to risperidone response. Furthermore, the prior studies looked at SNP associations, while the MIND team studied gene expression signatures.

One gene that the MIND team identified, RNF40, was notable because of its placement in a chromosomal region implicated in both autism and schizophrenia. Lit said the gene signature the group established wasn't the profile she would have predicted, "but it was interesting to see that, especially for one of the genes with changed expression, this is in a region that we already know is implicated in both autism and schizophrenia, so you might expect that it would be involved in the response," she said.

RNF40 and RNF213 both have RING domains, the authors wrote, containing a zinc-finger binding site, which was interesting because zinc levels while taking risperidone have been previously found to be associated with the level of behavioral improvement. Their findings, they wrote, "provide direction for further studies considering the relationships between expression of these RING-finger genes…and the relationships between zinc status and risperidone response.

Lit stressed that the findings will require significant validation to confirm the association between gene signature and risperidone response. She said the researchers are "continuing work along these lines."

"Of course I can't emphasize this enough, but these kinds of studies have to be replicated. You can't draw any absolute conclusions from a single study," she said.

In the study, the authors wrote that they are conducting additional research to refine their understanding of the relationship "between gene expression and adverse effects," and the potential influence of dosage on the selection of genes in their initial signature. In this first examination, variations in dose may have influenced the genes selected for further examination, by skewing the pattern of highest and lowest responders.

The authors also wrote that future studies will be necessary to determine whether the group's profile predicts risperidone in any other disorders, like schizophrenia, or with other antipsychotic drugs used in ASD. Lit said she wasn't aware of any contact between the team and pharmaceutical companies. But, she said, developing a commercial predictive drug-response test for ASD is certainly a goal of the MIND Institute researchers.

"Absolutely, the ultimate endpoint would be something that we could provide that would allow the opportunity to say, 'Yes, this child is probably going to respond to this drug,' or, 'You know what, maybe we need to just try another drug'," she said. "That would just be wonderful."

Robert Hendren, executive director at the MIND Institute and an author on the study, said in an email to PGx Reporter that he has been "negotiating with [Bristol Myers Squibb] about [supplying] drug and placebo if we get funded to do a similar study with aripiprazole," which is marketed by BMS as Abilify and is FDA approved for the treatment of irritability associated with autism in children between six and 17.

Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at mashford [at] genomeweb [.] com.

The Scan

Germline-Targeting HIV Vaccine Shows Promise in Phase I Trial

A National Institutes of Health-led team reports in Science that a broadly neutralizing antibody HIV vaccine induced bnAb precursors in 97 percent of those given the vaccine.

Study Uncovers Genetic Mutation in Childhood Glaucoma

A study in the Journal of Clinical Investigation ties a heterozygous missense variant in thrombospondin 1 to childhood glaucoma.

Gene Co-Expression Database for Humans, Model Organisms Gets Update

GeneFriends has been updated to include gene and transcript co-expression networks based on RNA-seq data from 46,475 human and 34,322 mouse samples, a new paper in Nucleic Acids Research says.

New Study Investigates Genomics of Fanconi Anemia Repair Pathway in Cancer

A Rockefeller University team reports in Nature that FA repair deficiency leads to structural variants that can contribute to genomic instability.