TORONTO – Current standards to help diagnostic laboratories and clinicians classify sequence and copy number variants with a potential role in Mendelian diseases are getting a makeover.
At the American College of Medical Genetics and Genomics annual meeting here on Tuesday, members of two working groups developing new versions of sequence variant and CNV standards provided a glimpse of the proposed changes and asked participants for their opinions. Both working groups, working in parallel and on slightly different timelines, represent ACMG, the Association of Molecular Pathology (AMP), the College of American Pathologists (CAP), and the Clinical Genome Resource (ClinGen).
The sequence variant working group has been revising the existing 2015 guidelines for several years now and is "pretty close" to finalizing a draft, according to Heidi Rehm, chief genomics officer at Massachusetts General Hospital and a member of the group, which is chaired by Les Biesecker of the National Human Genome Research Institute and Steven Harrison, laboratory director at Ambry Genetics. "We want to be as prescriptive as possible and as objective as possible," she said.
Almost 90 laboratories and clinicians have already signed up as pilot users, she added, and the working group is keen for more groups to give them a test drive.
Starting with the coming update, called SVC (sequence variant classification) v4.0, new versions of the standards will be numbered and older versions will be assigned a number retrospectively — the 2015 version, for example, will be SVC v3.0. Also, rather than coming out with a new version every five years or so, the working group plans to release interim updates in the future, Rehm said.
Like the existing standards, SVC v4.0 will help determine whether or not a variant in a gene that already has an established role in a Mendelian disorder is pathogenic. It will only apply to variant classification, though, and not to the clinical assessment of a variant's role in a specific patient, which will need to take the phenotype into account.
Rehm said that revisions for SVC v4.0 aim to address the appropriateness of current criteria, resolve ambiguities in their application, look at the strength assigned to criteria, and improve guidance for applying both pathogenic and benign criteria. Further, it will incorporate updates from ClinGen's Sequencing Variant Classification Working Group and, where possible, develop more quantitative measures. Like before, multiple lines of evidence will be combined to classify a variant, but the new version will change the structure for considering the evidence.
Harrison explained that SCV v4.0 will continue to use codes for different types of evidence and will assign them point values. The final number of points will determine how a variant is classified, ranging from 10 or more points for a pathogenic variant to negative 4 or fewer points for a benign variant.
Under the new standards, variants of uncertain significance will also be divided into three subcategories — low, medium, and high — based on the number of points assigned to them. While labs will not be required to use or report those subcategories, the hope is that they will track them internally to see which VUS become reclassified over time. A separate ACMG committee is working on guidelines for reporting VUS, he added.