Personal genome can predict abdominal aortic aneurysm risk

Last Updated: 2018-09-10

By Will Boggs MD

NEW YORK (Reuters Health) - An individual's genome, coupled with electronic health record data, can accurately predict the risk of abdominal aortic aneurysm (AAA), researchers report.

"I think genome sequencing will become a standard part of clinical care in the future," Dr. Michael Snyder from Stanford University School of Medicine, in Stanford, California, told Reuters Health by email. "From this study, we can use it to predict risk of a fatal disease. The approach can be applied to other diseases to help reveal their genetic underpinning as well as potentially lead to clinical tests."

Many complex diseases have a strong genetic component, though most cannot be linked to a single gene. Identifying the genomic propensity for a particular disease might foster guidelines on lifestyle adjustment and other steps that could be taken to minimize disease risk, Dr. Snyder and colleagues note in Cell, online September 6.

As a proof of principle, the team aimed to identify genomic variants associated with AAA. They performed whole-genome sequencing for 268 AAA cases and 133 controls and used machine learning and network analysis to identify genes and biological pathways associated with the disease.

No single genomic locus was significantly associated with AAA. The hierarchical estimate from agnostic learning (HEAL) analytical framework, however, identified a minimal set of 60 genes whose mutational burden was elevated in cases relative to controls.

The framework based on these genes predicted AAA status with 69% accuracy. Adding electronic health record data improved the predictive power to 80%.

The 60 HEAL genes showed an overall enrichment for immune-related functions, consistent with known immunological/inflammatory features of AAA pathophysiology.

In further analyses, functional modules associated with these genes showed significant upregulation of proteins previously associated with abnormal systemic arterial blood pressure and dilated-cardiomyopathy phenotypes.

"We can predict someone's risk of a fatal complex disease right from their genome sequence," Dr. Snyder said. "That means we can know right from birth what their risk score is. The risk score is improved if we add additional lifestyle and clinical information. In addition to screening for the disease, it gives them the chance to improve their lifestyle."

"I think we will get a better handle on the genetic underpinning of many common diseases," he said. "Someday people will be getting their genomes sequenced before birth or shortly afterwards and this will become part of their clinical care."

Co-author Dr. Philip S. Tsao, also of Stanford, told Reuters Health by email, "We would only expect this approach to be more valuable as more and more of the ancillary data in regards to protein-protein interactions and clinical data is collected, annotated, and made available to help manage disease, even before disease overtly manifests."

Five of the nine authors are listed as inventors on a pending patent related to the HEAL framework. Two of these authors also have relationships with one or more "omics" companies.

SOURCE: https://bit.ly/2Qhe3Rb

Cell 2018.