Genomic-First Approach Offers Potential for Earlier Diagnosis of Rare Disorders

A "genomic-first" approach to screening for rare genetic disorders -identifying specific genetic variants and then studying associated traits and symptoms - can identify these conditions earlier and more frequently than standard genetic testing driven by clinical symptoms, a Geisinger study found.

Rare genetic disorders (RGD) affect at least 24 million people in the United States, or more than 5% of the total population. Diagnosis of RGD has historically resulted from a "phenotype-first" approach, in which people with clinical symptoms are referred for genetic testing to identify an underlying cause. This approach can underestimate the prevalence of RGD, as people with less severe symptoms, or no symptoms at all, are less frequently referred for genetic testing.

For their study, the Geisinger team developed a list of 2,701 RGDs that are not routinely screened at the population level and then created a strategy for identifying disease-causing variants in this gene list within a group of 218,680 participants in Geisinger's MyCode Community Health Initiative.

The research team developed and applied automated methods for comparing participants' genomic findings to existing clinical diagnoses, which they defined as "diagnostic fit" (DxFit). They discovered that 2.5% of this group had a high-confidence genetic change for an RGD, but the DxFit assessment revealed that the majority of these people did not have evidence of a corresponding clinical diagnosis in their electronic health record.

The study was published online this week in the American Journal of Human Genetics.

This important finding suggests that using a genomic-first approach can identify many more people with rare disorders earlier and may also mean that the chance of getting sick from these genetic changes is lower than previously thought. Using a genomic-first approach offers the potential for earlier and more precise diagnosis, improved management and treatment, and a more accurate description of the symptoms of rare genetic disorders, all of which could contribute to improved outcomes."

Kyle Retterer, MS, chief data science officer at Geisinger and senior author of the study