NEW YORK (GenomeWeb) – A new clinical test that identifies disease-specific epigenetic signatures in DNA from blood samples promises to diagnose neurodevelopmental disorders that could not be solved by genetic testing and to replace targeted tests for imprinting disorders.
The genome-wide DNA methylation test was developed by a Canadian team led by Bekim Sadikovic at the London Health Sciences Centre. According to the researchers, it is the first test of its kind to become clinically available, though other groups are working on similar assays.
Greenwood Diagnostics Laboratories at the Greenwood Genetic Center in South Carolina will start offering the test in the US next week, while the Genome Diagnostics Laboratory at the Amsterdam University Academic Medical Center in the Netherlands plans to offer it in Europe beginning in June.
Last week, the LHSC researchers published a clinical validation study of the test in the American Journal of Human Genetics. In that study, they showed that DNA methylation signatures, or epi-signatures, can distinguish between more than a dozen genetic disorders that are caused by mutations affecting a cell's epigenetic machinery. Sadikovic, who is head of molecular genetics at LHSC and an associate professor at Western University, plans to present his work at the American College of Medical Genetics and Genomics annual meeting in Seattle this week.
The new publication follows a study in the same journal last year that demonstrated the ability of DNA methylation profiles to identify patients with a number of specific genetic conditions. In the meantime, the researchers published additional papers, including one in Nature Communications last November on the utility of epi-signatures for diagnosing BAF-related neurodevelopmental syndromes.
In their latest report, they focused on diagnoses made with epi-signatures after other test technologies such as exome sequencing or microarrays provided no answer or came up with a variant of unknown significance. According to Sadikovic, the additional diagnostic yield is on the order of 3 to 5 percent, a number he expects to grow as more conditions with characteristic epi-signatures are being identified.
"The paper shows that one can use this methylation profiling approach to distinguish cases that have pathogenic mutations (even where the exact mutation is not known) from those that carry benign variants," said Andrew Sharp, a researcher in the Department of Genetics and Genomic Sciences at Mount Sinai School of Medicine who has studied epigenetic variations in congenital disorders. "Thus, methylation profiling does have the potential to identify patients with pathogenic mutations that are often missed by exome or genome sequencing," he said.
The new test might also be useful for diagnosing imprinting or repeat expansion disorders with altered methylation patterns, such as fragile X syndrome, the most common form of intellectual disability in males. These are currently diagnosed using a variety of targeted assays that could potentially be replaced by a single DNA methylation test, Sadikovic said.
The test, which Greenwood Diagnostic Laboratories calls "EpiSign," runs on an Illumina Infinium MethylationEpic BeadChip array and currently covers 19 different conditions, including Down syndrome, Kabuki syndrome, and Cornelia de Lange syndrome. Using a patient reference database the LHSC researchers have built over many years, as well as machine learning, Sadikovic's team developed an algorithm that can identify epi-signatures in the methylation data.
Greenwood has licensed the patented knowledgebase and algorithm technology exclusively in the US, where it will provide the EpiSign test. While Greenwood will be in charge of marketing, methylation data generation, reporting, and billing, LHSC will analyze the data using its reference database and algorithm.
The arrangement with the University of Amsterdam's AMC for the European market will be similar, Sadikovic said, and the plan is to launch the test there in the next few months, around the time of the European Society of Human Genetics annual meeting in June.
"There are other labs we're working with right now that are interested in doing something very similar for their jurisdictions," Sadikovic said.
Michael Friez, director of the Greenwood Diagnostics Laboratories, said that his group has been collaborating with the Sadikovic lab for several years, initially providing samples from patients with a variety of clinical conditions to test the DNA methylation platform. "It really evolved from taking a research exploratory look at these different conditions with genes that are involved in DNA methylation activities," he said. "Once the list grew, it became apparent that we should be thinking about trying to market this as a test, because it would have a variety of different applications to different clinical populations."
One use of the EpiSign test, which will have a list price of $1,200, will be for patients where genetic testing has identified a variant of unknown significance. "We can help interpret whether these VUSes are pathogenic or not," he said. "That's an ongoing clinical dilemma for a lot of patients with a variety of conditions."
The second indication will be for imprinting and repeat expansion disorders, and that version of the test will have a list price that starts at $1,500. The plan is to offer the EpiGen test instead of individual tests for conditions like fragile X syndrome, Angelman syndrome, and Prader-Willi syndrome, which Friez said are commonly ordered in the early phase of evaluating a patient with intellectual disability. "So we're looking at the option of doing this as just one comprehensive test rather than multiple individual tests," which would be more efficient and less expensive, he said.
Friez said he expects the EpiSign test to be covered by insurance "in some cases" because tests for several of the conditions it covers already have CPT codes that can potentially be stacked together. In other cases, the lab will use a miscellaneous CPT code that is used as a catch-all. "Ideally, we would like to move towards this particular test, or type of test, having specific recognition for insurance coverage, and perhaps having its own CPT code for billing purposes," he added.
Marcel Mannens, head of the Genome Diagnostics Laboratory at Amsterdam UMC, said that his group has been studying the role of epigenetics in disease for many years, so "it was a logical thing to join forces and combine research efforts and diagnostic expertise" with the LHSC team.
"We think the EpiSign test has great potential for diagnostics," he said, as it covers 20, and soon will cover 40, disorders in a single microarray experiment. It can also serve as a functional test for VUSes and to diagnose imprinting disorders. "I can imagine that a combined CGH/methylation array, together with an exome or genome sequencing test, will be the default test in the near future in a large population of patients," he said.
However, "before we get there, we need larger databases of EpiSign [test data] and less expensive technology," he said. Right now, because of low sample numbers and expensive infrastructure, the price for the test will be around €1,070 ($1,200), he said, but "this can change fast if industry and healthcare recognize the potential of this new technology."
Because no other clinical genome-wide DNA methylation tests currently exist, it will be important to develop clinical guidelines for their use. "We expect that if this technology takes off and there is a demonstrated long-term clinical utility, there will probably be more laboratories doing this," Sadikovic said. "We're now the only ones doing it, technically speaking, so we have developed our own standards and guidelines, in particular since we're now trying to utilize the technology across multiple countries and laboratories," he said. However, when similar tests come to market, it would be helpful for organizations such as the American College of Medical Genetics and Genomics to put guidelines in place.
Sadikovic said a lot of work in epigenetic signatures has been done in cancer, an area his group is also interested in, but several groups besides his own have developed DNA methylation signatures for patients with constitutional disorders. "I would be surprised if there would not be a lot more of this in the coming years," he said.
"There are a number of groups that are doing very similar work that, to our knowledge, don't appear to be in a position or ready to launch it commercially, or at a clinical level," Friez said. "But there are a number of groups that have demonstrated that they are working on very much the same thing."
Going forward, the adoption of epigenetic testing could be similar to that of microarray-based copy number variation testing, Sadikovic said, which replaced targeted FISH assays for microdeletion or microduplication testing over time. "It was essentially a comprehensive assay to replace about a dozen of these different specific targeted analyses," he said. Also, just like the clinical use of microarrays led to the discovery of new microdeletion and microduplication syndromes, the clinical adoption of DNA methylation testing could uncover new epigenetic disease signatures. "In a clinical setting … with thousands and thousands more of these datasets, we should be able to discover a lot more utility," he said, adding that his lab will continue to expand its reference database.
"This whole world of methylation has really gotten a lot of attention over the last couple of years," Friez said. "It's not necessarily new technology but it's certainly a new way of applying informatics expertise to the data."