By Kevin Davies
December 13, 2012 | A small biotech company, Population Diagnostics (PDx), has collaborated with researchers at The Hospital for Sick Children in Toronto (SickKids) to provide a potentially major step forward in the search for a diagnostic test for autism.
Writing in the journal G3 (Genes│Genomes│Genetics), the researchers describe the discovery of a handful of gene variants for autism spectrum disorder. The variants are said to confer clinical utility and could spur the development of early detection diagnostic tests as well as aid in the quest for new drug targets for the disorder.
The collaboration, led by Stephen Scherer at SickKids, sought to understand microarray platform differences in copy number variant (CNV) detection. With the help of PDx, which delineates benign and pathogenic CNVs, Scherer says his group was able to interrogate the genome at a higher resolution than had been possible with other forms of microarrays, thereby detecting smaller variants associated with autism.
“Not only were we able to confirm variants in genes that we and others have previously discovered using alternate methods, but more importantly, we are pleased that an abundance of novel variants have been uncovered,” said Scherer. “We intend to report on additional discoveries in future publications that are based on an even finer-scale interpretation of the data in this joint project.”
Co-founded in 2006 by CEO Jim Chinitz, PDx holds a pair of patents on a method of CNV comparison in patients compared to healthy controls, spotlighting those deemed to be associated with the particular disease.
Chinitz told Bio-IT World that his company is driven by two key concepts: First, rare variants cause common, complex diseases. Second, a survey of rare CNVs in a disease cohort compared to a large number of healthy controls can pinpoint disease-causing genes, which are subsequently identified via targeted sequencing.
“Essentially, we have developed a proprietary core technology that quickly and cost effectively uncovers rare variants using any well-characterized cohort,” says Chinitz. “A cohort does not have to be a disease, but it could also be users of a drug who are responders, non-responders, or patients who experienced a serious adverse event.”
“What is under-appreciated is that CNVs are key for dramatically reducing the genome search space because there are far fewer CNVs to interrogate and interpret relative to SNVs,” says Chinitz. “Our gene discovery studies can be performed more quickly and inexpensively than rare variant study designs that employ only sequencing methodology… What sets us apart is that we actively developed a methodology for discovering [CNVs] and we can generate results without having to wait for next-generation sequencing to fully mature.”
In addition to discovering 16 novel genes associated with autism (including several implicated in neurodevelopment) in the new G3 paper, Scherer and colleagues point out that 64% of the CNVs observed were previously missed in work using single nucleotide polymorphism (SNP) microarrays. Because most of the CNVs skipped using SNP microarrays are relatively small, the new study has greater power to incriminate specific single genes involved in autism.
The G3 study underscores the notion that the genetics of autism, “involves numerous genes containing many low frequency genetic variants with large effect,” commented Peggy Eis, chief technology officer at PDx. The list of autism candidate genes, which will be confirmed and expanded in follow-up studies, represents “a significant portion of the unexplained genetic contribution to autism.”
Chinitz is skeptical about research based solely on gene expression signatures, which are dynamic measurements, he says, and thus prone to variation between individuals depending on diet, medications, time of day, and so one. Genomic DNA testing, by contrast, is a static measurement. “It remains to be seen whether an expression signature can be used as a biomarker to prospectively distinguish autistic children from non-autistic children using data retrospectively from patients diagnosed with autism,” says Chinitz.
The PDx test will stand out from its competitors because it will contain novel content and improve on the 10-15 percent diagnostic yield of autism tests currently on the market. “The novel content is obtained via higher resolution analysis of CNVs and subsequent gene sequencing to identify the full mutational spectrum of rare variants in autism-associated genes,” Chinitz said.
In addition to research on autism, PDx has active discovery programs in Parkinson’s disease, Alzheimer’s disease, peanut allergies, endometriosis, schizophrenia, and PML (progressive multifocal leukoencephalopathy, a myelin disorder), all of which are being pursued via other collaborative relationships.
Chinitz says PDx is platform-agnostic, and that the combination of “platform choice, disease choice and bioinformatics” together with the firm’s insights into CNV distribution are all mission critical to finding disease causing genes.
Chinitz co-founded PDx after stints with Hoffmann-La Roche, Applied Biosystems, Affymetrix, Third Wave Technologies and Enzo Life Sciences, largely in the fields of DNA sequencing, the polymerase chain reaction, and microarray technologies. He says his long-term goal is to discover as many new genes as possible for the pre-symptomatic and comprehensive diagnosis of patient subtypes for common diseases and to predict drug efficacy and safety.
FURTHER READING: Prasad, A. et al. “A discovery resource of rare copy number variations in individuals with autism spectrum disorder.” G3 December 1; 2, 1665-1685 (2012). http://www.g3journal.org/content/2/12/1665.full