Study Shows Benefits of Large-Scale, Clinical Sequencing Initiatives
Large-scale sequencing initiatives performed in integrated health care systems, such as the DiscovEHR collaboration, can advance genetic discovery and serve as a "blueprint" for adoption of precision medicine, according to a study published Dec. 23, 2016 in Science. The DiscovEHR study is a collaboration between the Regeneron Genetics Center (wholly-owned subsidiary of Regeneron Pharmaceuticals) and Geisinger Health System (Danville, Penn.). The initiative couples high-throughput sequencing and longitudinal electronic health records (EHRs) in a real-life clinical cohort that enables returning and acting on reportable variants. Sequencing of the exomes of 50,726 adult participants in the DiscovEHR study (sequence coverage of at least 20× haploid read depth at more than 85 percent of targeted bases in 96 percent of samples) identified roughly 4.2 million rare, single-nucleotide variants and insertion/deletion events, of which approximately 176,000 predicted a loss of gene function. Each individual had a median of 21 rare variants predicted to result in a loss of gene function. "Although many of these variants are rare individually, in aggregate, they are not uncommon, and their identification and the biological insights gleaned are relevant to our understanding and treatment of both common and rare diseases." —David Carey, Ph.D. The researchers found that linking these […]
Large-scale sequencing initiatives performed in integrated health care systems, such as the DiscovEHR collaboration, can advance genetic discovery and serve as a "blueprint" for adoption of precision medicine, according to a study published Dec. 23, 2016 in Science.
The DiscovEHR study is a collaboration between the Regeneron Genetics Center (wholly-owned subsidiary of Regeneron Pharmaceuticals) and Geisinger Health System (Danville, Penn.). The initiative couples high-throughput sequencing and longitudinal electronic health records (EHRs) in a real-life clinical cohort that enables returning and acting on reportable variants.
Sequencing of the exomes of 50,726 adult participants in the DiscovEHR study (sequence coverage of at least 20× haploid read depth at more than 85 percent of targeted bases in 96 percent of samples) identified roughly 4.2 million rare, single-nucleotide variants and insertion/deletion events, of which approximately 176,000 predicted a loss of gene function. Each individual had a median of 21 rare variants predicted to result in a loss of gene function.
"Although many of these variants are rare individually, in aggregate, they are not uncommon, and their identification and the biological insights gleaned are relevant to our understanding and treatment of both common and rare diseases." —David Carey, Ph.D.
|
The researchers found that linking these data to EHR-derived clinical phenotypes for more than 39,000 patients, uncovered associations supporting therapeutic targets, including genes encoding drug targets for lipid lowering, and identified previously unidentified rare alleles associated with lipid levels. Clinical records covered a median of 14 years and captured a median of 87 clinical encounters, 658 laboratory tests, and seven procedures per participant.
During consent, participants agreed to be recontacted for return of clinically actionable results to inform their health care. The researchers found that roughly 3.5 percent of individuals harbored deleterious variants in 76 clinically actionable genes and potentially pathogenic variants known as the Geisinger-76, which include the 56 genes and 25 conditions recognized in the American College of Medical Genetics and Genomics reporting recommendations plus an additional 17 genes associated with the same 25 conditions and three genes associated with two additional conditions.
Following expert clinical review and adjudication of potentially pathogenic variants, 43 reportable variants (all either heterozygous for autosomal dominant conditions or hemizygous for X-linked conditions) were identified in 49 individuals. The expressivity of these variants were investigated through reviewing EHR-derived clinical phenotype data for variant carriers. Nearly two-thirds (32 of 49 individuals) had documented clinical features in the EHR consistent with the associated disease, however only seven individuals (14 percent) had a formal diagnosis of the associated disease. Geisinger has stated that for the 200 patients already informed they carry one or more actionable, disease-causing genetic mutations, most variants are related to cancer risk and cardiovascular illness.
"Although many of these variants are rare individually, in aggregate, they are not uncommon, and their identification and the biological insights gleaned are relevant to our understanding and treatment of both common and rare diseases," writes senior author David Carey, Ph.D., from Geisinger Health System, in the study. "These results demonstrate the potential for genomics-guided clinical care in a large unselected clinical population, establish an expectation for the burden of actionable genetic findings, and support the need for expert clinical review and adjudication of assertions of pathogenicity for potentially pathogenic variants, including those cataloged in mutation databases."
Takeaway: Large-scale sequencing initiatives linked to clinical data can inform the practice of precision medicine and drive genomics-guided therapeutic discovery.
Subscribe to Clinical Diagnostics Insider to view
Start a Free Trial for immediate access to this article