Whole-genome sequencing has just moved a little bit closer to the average doctor’s office. The New England Journal of Medicine, probably the world’s most influential medical journal, has published a paper—a primer, really—that instructs clinicians in the basics of what authors Leslie Biesecker and Robert Green are calling Clinical Genome and Exome Sequencing.
CGES can involve an entire genome or the much smaller exome, the 1-2 percent of human genes that code for proteins. Biesecker, who is at the National Human Genome Research Institute, says thousands of patients have already undergone CGES, and he estimates that 10,000 more will this year. Green, who is at Brigham and Women’s Hospital and Harvard Medical School, calls it “a transformative moment in the history of medicine.”
That may sound like enthusiasm, but the paper is not an endorsement of the extensive personal genomics we’ve all been promised. Instead it emphasizes the very limited circumstances where clinical genomic sequencing makes sense today. Before they even begin talking specifics, Biesecker and Green make clear they don’t approve of sequencing healthy people. They note also that results from CGES are likely to disappoint. The track record will doubtless improve in future, but right now CGES claims only a 25 percent success rate.
Genome sequencing for rare disorders due to single genes
The authors recommend CGES only for detecting rare DNA variants in patients, usually children, who look as if they might be suffering from a mendelian genetic disorder. That’s a disorder due to a specific gene, one that is still a mystery after known single-gene disorders have been eliminated.
They also list the large number of conditions not easily detected by CGES. These include repetitive DNA (as in Huntington’s disease), copy-number variants (as in DiGeorge syndrome), abnormalities of chromosome number (as in Down syndrome), and several other kinds of DNA variations associated with disease.
Moreover, they say, an order to perform sequencing should be preceded by a lot of preliminary spadework: a very thorough family history, systematic evaluation of the patient’s phenotype, searching medical literature and databases for clues, and obtaining informed consent. This prior work might also involve sequencing other family members.
And even in the 1 in 4 cases when CGES reveals a DNA variation, figuring out whether the variant has clinical meaning is tough. It means searching databases for other examples and scouring the medical literature for clinical associations. And watch out; Biesecker and Green caution that the literature is full of errors and false attributions.
In short, they are proposing an in-depth approach, not anything like the kind of sequencing consumer gene testing companies do. Nor is this paper about the sort of whole-genome sequencing of ostensibly healthy people that I wrote about here at GLP in March.
What Biesecker and Green are describing is a very limited and yet complex process with many steps and some ethical issues, such as whether to disclose information to collateral relatives. Furthermore, for now CGES is potentially useful in only a small proportion of mystery ailments.
Share this article
In a very few cases, a CGES result will lead to treatment that makes a big difference. But most of the time even an accurate result will not mean direct therapeutic help, although it can still be useful. For example, it may put an end to the endless quest for diagnosis, or assess whether the condition might occur in future children, or lead to health information that other family members can use for treatment or prevention.
CGES is expensive and insurance coverage uncertain
And then there’s the cost, which Biesecker and Green estimate at between $4,000 and $15,000 per patient. It’s not clear whether insurance companies will be embracing CGES wholeheartedly.
In some cases they have paid, Biesecker and Green say. But at Reuters, Julie Steenhuysen reports that “a number of insurers, including Blue Cross Blue Shield, have reacted by putting the brakes on reimbursement, according to interviews with researchers, diagnostic experts and insurance executives. Insurers are demanding proof that the results will lead to meaningful treatments.”
And how will insurers define “meaningful treatments?” Biesecker and Green say finding genomic variation will not often lead to direct therapy for the patient. I suppose insurers might be enthusiastic about ending what Steenhuysen calls “hugely expensive medical odysseys” involved in seeking a diagnosis for a mystery disease, and perhaps they would consider information that bears on future reproductive decisions of the insured to be cost-effective too. But it’s a gamble with only a 1-in-4 chance of success at the moment, and no guarantee that even finding variant DNA will leave docs and their patients any more informed about the nature of a rare, enigmatic medical condition.
Steenhuysen reports on specific cases where insurance companies have declined to pay for tests. The article also quotes lab heads as saying that, as the tests have gotten more common, reimbursement has increasingly been denied.
A number of small labs are reportedly interested in taking on clinical genome sequencing but are waiting to see what insurance companies decide to do. Seems to me that might mean a traffic jam, for a while at least, at the limited number of labs that are doing CGES.
Tabitha M. Powledge is a long-time science journalist and a contributing columnist for the Genetic Literacy Project. She also writes On Science Blogs for the PLOS Blogs Network. Follow her @tamfecit.
