By most appearances, Eliana Lewis was a healthy baby girl at her birth on December 31, 2013. She arrived a month early at North Kansas City Hospital, a tad small and hypoglycemic, so doctors whisked her to the neonatal intensive care unit (NICU). That’s where the seizures started.
At first, the medical team thought her difficulties were somehow related to the low blood sugar. But after two days, when Eliana wasn’t improving, doctors transferred her to Children’s Mercy Hospital in Kansas City. Her parents had never held their daughter. “You have so many expectations for what birth is supposed to be,” says her mom, Michelle. “I knew she might have to spend time in the NICU, but when the transfer happened I thought, ‘Whoa. This is something more serious.’”
Still, the medical consensus was that Eliana would probably go home to her own crib in a couple of weeks. Two weeks passed, then a month, and finally the seizures became so serious that doctors put Eliana in a medically induced coma to protect her. At a loss to explain the baby’s symptoms, doctors ordered full genome sequencing. The results could take about a month.
When researchers asked whether Eliana’s parents would be interested in a faster test, one that could provide answers in a couple of days, they readily agreed. “For us, there was no downside,” Michelle Lewis says. “It gets to the point where you hope something comes back abnormal so you can have an answer.”
A Faster Test
Eliana’s test was a means of rapid genome sequencing being developed largely by Stephen Kingsmore, the executive director of medical panomics at the Center for Pediatric Genomic Medicine at Children’s Mercy. Genetic abnormalities are one of the most common causes of death and suffering in newborns. Sadly, the majority of known abnormalities have no treatments, even if a diagnosis is made quickly. But some genetic diseases and disorders in infants, if caught early, can be either minimized or cured. Most sequencing available now is ploddingly slow, causing doctors to lose valuable time when they could otherwise be limiting a baby’s distress or even curing the disease. Kingsmore is one of the few researchers trying to speed the time into mere hours. The time from blood sample to diagnosis is now about two days, but he believes he can soon compress it into around 24 hours, and eventually 18.
Most sequencing available now is ploddingly slow, causing doctors to lose valuable time when they could otherwise be limiting a baby’s distress or even curing the disease.
Even if he meets that goal, complications remain. Right now, the sequencing costs about $20,000 — too expensive to be offered routinely. Medical ethicists also point out that revealing an entire genome at birth can introduce legal and ethical implications that medical science is still working to understand. Nonetheless, there is a need: About 14 percent of newborns every year are admitted to a neonatal intensive care after birth, Kingsmore says, and of those, around 5 percent will, like Eliana, be severely ill.
“These are very sick babies and time is of the essence,” says Tiina Urv of the Eunice Kennedy Shriver National Institute of Child Health and Human Development. “For some children, it could take years to get a proper diagnosis, time the children don’t have. The more information you get and the faster you get it, the better it is for the child.”
For example, some conditions are metabolic malfunctions that interfere with a baby’s ability to process key nutrients. Given a standard diet, these infants could become profoundly mentally impaired or die. Although standard newborn screening will catch most of these conditions within a few days after birth, other diseases may only be identified with sequencing.
“Depending on which source you cite, there are more than 5,000 different genetic diseases,” Kingsmore says, and almost all of them show signs at birth. The conditions he’s talking about are rare, but can be gravely serious. Without genetic insight, a diagnosis must be made through a time-consuming and stressful process of elimination, treating symptoms and narrowing down the cause.
Hospitals in most states screen for 31 genetic diseases and 26 secondary disorders, but some states look for more. The problem is, routine testing only covers a fraction of known genetic conditions, says Anastasia Wise of the National Human Genome Research Institute. Analyzing the entire genome, when it is done at all, typically occurs after a baby is clearly ill.
Even with a genetic explanation in hand, a lot of babies will still die, Kingsmore says. “To get to that decision point quicker allows the family time to come to terms with what they may be feeling — guilt, anxiety, regret.”
The project in Kingsmore’s lab is just one part of a larger, $25 million government-funded project exploring the role of genome sequencing in the care of newborns. Other research efforts are underway at Brigham and Women’s Hospital in Boston; the University of California, San Francisco; and the University of North Carolina at Chapel Hill. NIH officials say the Children’s Mercy project is the farthest along with efforts to speed diagnosis through rapid whole genome sequencing.
It wasn’t a research direction Kingsmore had planned. In 2011, he says, a colleague in England, who works for sequencing software company Illumina Inc., asked out of the blue whether he had any interest in developing rapid sequencing in the newborn population. The technology works by slicing a DNA sample into pieces that are about 200 to 500 base pairs long (base pairs being the “letters” that make up the coding of the genome). A computer then splices those pieces back together and hunts for changes that are known to be associated with disease. The genetic variations are then compared with the infant’s symptoms.
Still unclear is the degree to which the test can save lives. In 2012, in the journal Science Translational Medicine, Kingsmore reported the results of what was then a 50-hour genome sequencing proof-of-principle study. (For comparison, turnaround time for other genetic sequencing tests can be about four to six weeks.) Part of the study included testing four infants who were suspected of having genetic disorders but had not yet received a diagnosis. A specific genetic cause was identified for three of them. While these findings represent a step forward in diagnosis, by the time the study was published, three of the four babies had died. Even with a diagnosis, a treatment or cure is sometimes unavailable.
A study Kingsmore published in an April 2015 edition of The Lancet Respiratory Medicine paints a slightly more optimistic picture. The study found that rapid whole-genome sequencing identified the genetic disease in 20 of 35 critically ill infants. A diagnosis did alter the course of treatment for some patients in the form of palliative care or actual treatment options. However, 12 of the 20 infants whose diseases had been identified through rapid sequencing died.
Kingsmore and his colleagues hope to eventually enroll 1,000 newborns in the research project. All will receive conventional newborn screening, but about half of them will also get rapid whole-genome sequencing.
The concern with sequencing is that information parents want may come with information they don’t. Genetic information may reveal a condition an infant may be at risk for in the future or may reveal a previously unknown family risk. “Not only do you have this information about the child, but this information might also be relevant to their family. It could affect more than just the parents,” says Wise of the NIH. To limit the risk of accidental bad news, researchers are focusing their tests on 600 genes that are known to be associated with congenital problems. (They won’t, for example, look for mutations in the BRCA genes that are linked to breast cancer.)
That isn’t the only complication. Some children who test positive for genetic diseases might not ever develop symptoms, triggering unnecessary anxiety in parents, and perhaps leading to a treatment a child does not need. Testing could also create what researchers have called a “therapeutic gap,” revealing a genetic disorder for which no treatment exists.
Eliana’s parents were told their baby had a mutation in the SCN2A gene, which is involved in the function of the central nervous system. It’s rare and has no cure, though a special diet seems to help. Her parents were finally able to bring her home when she was 5 months old. Her mom says some days are better than others, and she knows there is little chance her daughter will ever be well. She and other parents have formed a foundation, FAMILIEScn2a, to encourage research.
“I can look back now and realize that having that diagnosis quickly gave us a chance to process,” Michelle says “We know what we’re fighting against now.”