Whole-genome scans: Promising too much, or too little?

Last week, I made a brief reference to a study in which doctors were able to scan the entire genomes of critically ill newborn infants in only two days, with an eye toward one day identifying and treating genetic abnormalities before they proved fatal.

The story received only modest coverage, except in The New York Times, which ran a lengthy story by Gina Kolata on page 1. The Times story begins with a feature lede about a baby girl who had repeated seizures from the day she was born. Every conceivable intervention was tried until, Kolata reports, "her family decided to let their baby go, and the medical devices were withdrawn. She was 5 weeks old."

I was unimpressed with the story at first, because it seemed to me to be theoretical: The treatment was a success, but the patient died. The researchers discovered the cause of the baby's illness, but they couldn't have done anything about it. Kolata does note that the study was intended to prove the concept, not treat the children, and that the results were not immediately available to the infant's doctors. Kolata quotes a lament by one of them: "If only…the test could have been done within days of the baby's birth" he tells her. But even if it had been done in time, it would not have saved the life of the tiny patient. "There was no treatment, there was not anything that could have changed the outcome," the doctor tells Kolata. The baby "was heavily sedated, medicated, and intubated for her entire brief life." In the study, the researchers tried the genome scans on four babies; only one survived. 

Perhaps I'm making too much of this, but Kolata did pin her story on the story of this baby girl. Did Kolata promise too much? She quotes another doctor saying "more research needs to be done before the test is ready for widespread use." I think Kolata's study provided a fair description of what happened, with the appropriate cautions and emphasis that the technique is not yet ready for routine use. 

Here is how some others covered the news:

Alan Bavley of The Kansas City Star (the research was done in Kansas City, at Children's Mercy Hospitals and Clinics) backs into the story, observing the importance of quick diagnosis before revealing the news, that researchers can now do in two days what has taken four to six weeks. He gives equal weight to the two possible results of such scans: Doctors can start treatments early, or suspend tests and treatments when there is no hope.

Jessica Stoller-Conrad at ShotsNPR's health blog, takes a much more cautious approach, reporting that "in-depth personalized genome sequencing could be inching closer to clinical reality." She also mentions another study from the New England Journal of Medicine in which researchers sequenced some of the DNA of 100 patients with severe intellectual disabilities. The researchers identified the genetic causes of the disability in 16 of the patients. As was the case with the newborn study, the diagnoses "don't necessarily come with an effective treatment," Stoller-Conrad writes.

Robert Langreth of Bloomberg reports that the study "brings genome mapping a step closer to everyday hospital use." And, as I wrote in my brief mention last week, Julie Steenhuysen of Reuters quotes a researcher who says, "We basically struck out completely, for now anyway."

Interestingly, all of these accounts were scooped by genomeweb.com, which had the story in July, from a webinar offered in June by Illumina, the maker of the sequencing machine used in the study. And the genomeweb story, by Molika Ashford, included information on nine patients, not merely the four in the published study. That story also said that the hospital plans to begin offering the gene-scanning service to patients later this year. (As an aside, Illumina's stock jumped about 6 percent, from about 47 to 50, between Sept. 28 and Oct. 2–just before the results of the study were published.) 

So did the coverage promise too much? We will know more when and if Children's Mercy begins to test newborns routinely. Until then, I'm going to wrap myself in the protective warmth of a cliché: Time will tell.

-Paul Raeburn

Leave a Reply

Your email address will not be published. Required fields are marked *