Proving The Value Of Preventive Genomics

By Deborah Borfitz  

October 15, 2020 | The Bio-IT World Conference & Expo closed out with a plenary keynote presentation on preventive genomics by Robert Green, M.D., professor of medicine at Harvard Medical School and a physician-scientist who directs the G2P Research Program at Brigham and Women’s Hospital and the Broad Institute. Data-sharing difficulties were a recurring theme at this year’s conference but, as the COVID-19 Host Genetics Initiative has demonstrated, it is possible to combine genomic data to rapidly explore markers of disease, he says. But far more daily deaths are caused by cancer and cardiovascular disease—not the pandemic virus—and 59 of the causal genes are already known and actionable.

Genomic information is rarely incorporated into clinical care partly because labs, not care providers, are doing most of the testing and doctors are unclear if the benefits outweigh the costs and risks, says Green. The clinical value of DNA sequencing is also unproven, although it’s the central feature of personalized medicine programs that have been popping up around the country.

Green presented lessons learned from the MedSeq, exploring the impacts of incorporating genomic sequencing into everyday medicine for people with and without a suspected genetic cardiac disease, and BabySeq, testing methods for integrating sequencing into the care of newborns. Both are randomized trials funded by the National Institutes of Health. 

MedSeq involved primary care physicians taking comprehensive family histories on participants with or without the addition of one-page genomic reports and following their outcomes. Reports from preventive genomic testing focused on defined, disease-specific variants with the “highest clinical actionability,” says Green, as distinct from indication-based testing looking at a wider universe of variants known or suspected of being pathogenic. 

Notably, Green says, neither doctors nor patients experienced test-related anxiety—even when a monogenetic risk variant was discovered. In 100 individuals, 20% were found to carry a dominant mutation for a monogenetic condition. In fact, among the top four genetic mutations, sequencing often discovered ongoing disease that the healthcare system had missed.

Participating doctors, after only six hours of training, did not make any errors in communicating the results, adds Green. Healthcare spending six months post-disclosure was higher but “not extraordinarily more.” Two years later, 22% had been reclassified (e.g., “variant of uncertain significance” now “likely benign” or “likely pathogenic” variant now “pathogenic”).

In the smaller BabySeq Project, 11% of participants were identified as having monogenetic disease risk, Green says. As with MedSeq, a substantial number with genetic mutations already had phenotypic evidence of disease previously missed by their healthcare providers. 

BabySeq additionally revealed no difference in bonding or vulnerability, says Green. “Catastrophic distress is not an obstacle [to sequencing],” as has often been suggested. The falling cost of genomic sequencing and interpretation should further improve the benefit-to-cost ratio.

Exactly how often does sequencing reveal something important? Here’re the stats from Green: 91% of the time for recessive mutations, 80% for atypical responses to medications, 15% for dominant mutation, and 50% for elevated polygenic risk specific to at least one condition such as diabetes or cancer. 

Polarizing Topic

The Mass General Brigham Biobank, which looked for the 59 genes linked to disease, has identified such mutations in over 350 of the roughly 36,000 people it has sequenced. In 75% if those cases, the mutations were linked to either cardiovascular disease or cancer and the individuals had no idea they were carrying mutations, says Green.

A significant number did not even want to know of their risk, he adds. A similarly high number met National Comprehensive Cancer Center criteria for genetic testing but had never before been tested.

The Preventive Genomics Clinic at Brigham and Women’s Hospital, staffed by genetics experts and counselors, offers individuals a menu of testing options (whole genome sequencing as well as smaller panels) and also gives patients the option of being seen via telemedicine. The heart-touching stories shared on its website include a man nudged by discovered mutations to finally get a colonoscopy, revealing two cancerous lesions that were subsequently extracted, and another with worsening heart disease who learned the underlying cause was Fabry disease—a rare but treatable condition.

Genomics is a notoriously polarizing subject, Green says. The challenge in convincing the skeptics is that genomics crosses multiple therapeutic domains and testing needs to be repeated over individuals’ lifetime. 

“The exceptionalism of genomics is sometimes misplaced,” he later adds, referring to the disproportionate amount of fear about misuse of genetic information relative to psychological or infectious disease data. “It’s perfectly possible for large groups to share genomic data that is not identifiable. It’s not full-proof, but it’s [technically] feasible.” 

Federal genetic privacy laws prevent genetics-based discrimination by employers and health insurers, Green says. In July, Florida became the first state in the nation to enact a DNA privacy law that also prohibits life, disability and long-term care insurance companies from using genetic tests for coverage purposes.    

Editor’s Note: Even if you missed the start of the event, Bio-IT World Conference & Expo virtual is still live. Register now for on-demand presentations.