Experts met at the US Food and Drug Administration (FDA) today to discuss how curated databases could be used to establish the clinical relevance of genetic variations, and how the agency could use these databases to evaluate next-generation sequencing (NGS) diagnostics.
NGS tests have the ability to quickly sequence the human genome and identify thousands of genetic variants. However, as Erin Ramos of the National Human Genome Research Institute put it, our "ability to detect genetic variants far surpasses the ability to interpret the phenotypic impact of variation."
FDA is considering ways for NGS test developers to leverage well-curated databases to help provide clinically meaningful or actionable results to patients.
Ahead of today's meeting, FDA released a discussion paper, Use of Databases for Establishing the Clinical Relevance of Human Genetic Variants, in which it discussed strategies to identify and implement standards to "ensure that NGS tests produce accurate and reliable results."
Because some genetic variations identified in patients with particular diseases occur infrequently, FDA recognizes that it may "not be possible to prospectively identify enough individuals with the same genotype to provide the necessary sample size to demonstrate the clinical significance of particular variants."
The solution FDA is proposing would be to create standards to aggregate and curate genetic databases.
"Our particular interest is databases as a source of clinical validation evidence for particular variants, groups of variants, haplotypes … so different test developers are not required to generate the data themselves. They do not have to go out and collect patients with this variant and look at their phenotype or their response to drugs. We think a lot of this is being recorded [in databases] already," FDA's Elizabeth Mansfield, deputy director of personalized medicine and molecular genetics said.
However, FDA has identified a number of issues and considerations for that must be taken into account before that can be done, including how to assure database quality, how to curate databases and how to communicate different databases interpretations of genetic variants.
Panelists disagreed over the need for specific standards for training database curators, or what education level or qualification these staff should have. But some agreed with Karen Raraigh, a genetic counselor at Johns Hopkins University, who emphasized the importance of having "multiple eyes, multiple hands" involved in the curation of any data.
Ramos concurred, adding that "it's really critical to have … a framework that is transparent that curators can understand."
Michelle Whirl-Carrillo, associate director of PharmGKB, expressed concerns that different database efforts don't use the same terminology, "even the vocabularies are not lining up … common terminology would be very useful, if we can ever get to that state."