Diagnostics

50,000-person genomics study reveals untapped medical potential

Results from a momentous 50,000-person genomics study were released on Thursday, generating rich data on the frequency of disease-causing genetic variants and how they present at the doctor’s office.

Urban crowd from above

Christmas came early for geneticists this year.

Results from a momentous 50,726-person genomics study were released on Thursday, with some interesting and actionable lessons that have many experts talking.

presented by

Published as two papers in Science, the research stems from the DiscovEHR project, led by researchers at the Geisinger Health System in Pennsylvania and the Regeneron Genetics Center, an off-shoot of Regeneron Pharmaceuticals.

The study combined whole-exome sequencing results with the longitudinal electronic health records (EHRs) of 50,726 individuals. The stated aim was to better understand common genetic variants, their frequency, and their link to human disease.

According to The Scientist, the researchers focused on a subset of 76 genes linked to 25 clinically-actionable disorders. The list included hereditary cancers and a set of variants linked to familial hypercholesterolemia — LDLR, APOB, and PCSK9.

atul-butte

Of the more than 50,000 participants, 3.5 percent harbored at least one gene variant linked to a disease. The number surprised Daniel Radar, a prominent geneticist who penned the accompanying editorial.

“This frequency of theoretically clinically actionable mutations is somewhat surprising and higher than one might have guessed in an unselected cohort of patients,” Rader told The Scientist.

Using the same patient cohort, the researchers did a deep-dive into familial hypercholesterolemia (FH), a hereditary condition that leads to early, aggressive cardiovascular disease.

Many undiagnosed cases of FH were found within the patient cohort for an overall prevalence of around 1 in every 256 patients.

The press release claimed that number to be twice as high as many prior studies had predicted, but it is consistent with the Familial Hypercholesterolemia Foundation’s statistics, which cite 1 in 250.

eric-topol

When it came to treatment, however, a lot was revealed. Only 24 percent of people with FH variants could be diagnosed with the condition based on electronic health records alone (the median participant age was 48). For the other 76 percent, the genetic test was critical for recognizing the condition and getting adequate treatment.

The study found 58 percent of patients with FH variants had a recent active prescription for statins, the go-to therapy for preventing cardiovascular complications. The rest did not. For those taking statins, less than half met the established goals for cholesterol lowering.

In his editorial, Rader paid tribute to the incredible scale and potential of the study.

One of the greatest challenges of the biomedical enterprise is to link human genetic variation with phenotypic traits at a population scale. Such efforts have myriad benefits, including the illumination of basic human biology, the early identification of preventable and treatable illnesses, and the identification and validation of new therapeutic targets, thus enabling the promise of precision medicine to improve human health.

A study co-author told The Scientist that an additional 75,000 patients have consented to join the study. The DiscovEHR team is also planning to expand the scope of the genes and diseases analyzed.

Photo: georgeclerk, Getty Images

Topics