With 10 more weeks until her baby was due, Sunnyvale mother Sici Tsoi’s ultrasound looked good, her doctors said. But her baby’s heart rate was slow. Her doctors sent her to a San Francisco hospital for a second evaluation, but the new medical team told Tsoi the same thing. The structure of the baby’s heart looked fine, but something was wrong with the heartbeat.
Scheduled for an emergency C-section, Tsoi checked into Lucile Packard Children’s Hospital Stanford, where doctors pulled baby Astrea from her mother’s womb and swiftly took the infant away.
Tsoi had two healthy daughters at home, and so she and her husband, Edison Li, assumed the doctors would make sure Astrea was fine.
Even hours after Astrea was born, Tsoi and Li waited patiently to hear how their new daughter was doing. They hadn’t even seen her, remembers Tsoi, and she and her husband just hoped for the best. “All I knew was that I gave birth, and I was very happy.”
Astrea was anything but well. Every few hours, she went into cardiac arrest, and it was all her doctors at Packard could do to keep her alive. The struggle to save her included a genetic analysis that eventually morphed into a study of genetic mosaicism, a condition in which some of a person’s cells carry different gene variants than the rest. Mosaicism could explain many undiagnosed conditions. In September, a team of researchers published their study of Astrea’s heart in the scientific journal Proceedings of the National Academy of Sciences.
Getting on the case
While Astrea’s parents waited calmly for word, Astrea was fighting for her life. Her doctors quickly diagnosed her with a heart arrhythmia called long QT syndrome. In long QT syndrome, a segment of the heart’s rhythm — the time between the start of the Q wave and the end of the T wave — is a little longer than normal. It’s often caused by a genetic defect and can be mild or severe.
For Astrea, the defect was severe. And, despite excellent care and treatment with different drugs known to help long QT syndrome, the medical team couldn’t keep her heart beating. They hoped that knowing the nature of the gene defect causing the problem would help them choose the right drug. But that would take time to discover and they didn’t have time.
When Tsoi’s doctors came to see her the morning after Astrea was born, she thought they were coming to check on her. She was surprised to learn that they were there to tell her that Astrea would soon be undergoing surgery to implant a pacemaker, which would regulate her heart rate better, as well as an implantable defibrillator, which would kick start her heart each time it stopped. Tsoi and Li were only now beginning to realize how possible it was that Astrea could die.
The pacemaker and defibrillator would give Astrea’s doctors time, but they knew it might not be a long-term solution. And knowing what kind of genetic defect was causing her heart to malfunction could help, too. Fortunately, just hours after Astrea was born, her case was discussed at a regular weekly meeting of Stanford Medicine’s Center for Inherited Cardiovascular Disease. Present were experts in cardiovascular genetics who could help, including Euan Ashley, FRCP, DPhil, and James Priest, MD.
“We realized how sick this child was,” Priest says, “and we had a new tool — rapid whole-genome sequencing — that could make a faster and more comprehensive diagnosis than the available clinical genetic testing. So that night I went and talked to the parents and the rest of the team, collected a blood sample and we started the test.”
On Astrea’s third day of life, Priest, an instructor in pediatric cardiology, looked at the DNA from her blood cells for any of several gene variants known to cause long QT syndrome. He found a suspicious mutation, but he immediately ran up against two problems.
First, he wasn’t positive that the mutation was capable of causing a heart problem as serious as hers. Second, the abnormal DNA was present in very small amounts. It didn’t make sense. Strands of DNA come in pairs, one from the mother and one from the father, so he’d expect half or even all of the DNA to have the mutation. Instead it was just 8 percent.
Because of that odd ratio, Priest wondered if Astrea might be a mosaic of two kinds of cells, normal and deadly. It was possible, he knew, for a mutation to have occurred when her cells were first dividing as an early embryo, just hours after conception, leaving a small lineage of related cells marked for a separate fate.
He suspected the deadly genetic mutation causing the problem might be hidden away deep inside Astrea’s heart.
But it would take a diverse group of nearly 30 researchers at many institutions, and a combination of some of the fastest and most cutting-edge genome sequencing ever conducted, to uncover the mystery of her illness.
Connecting the dots
Once newborn Astrea had her two implants, she spent six weeks at Packard gradually recovering from open heart surgery and undergoing more tests. But no one knew how long the implanted pacemaker and defibrillator would keep her alive and the genetics team scrambled to find out what was making her heart stop.
Connecting all the dots was a huge challenge. “It was two to three weeks of high-intensity work and about as dramatic as it gets,” says Ashley, a professor of medicine and of genetics who directs the Stanford Center for Inherited Cardiovascular Disease and co-directs Stanford Health Care’s Clinical Genomics Service.
First they had to confirm that the variant was real. They did this by turning to colleagues at a sequencing firm called Personalis, which Ashley had co-founded with other Stanford faculty. “The Personalis team dropped everything and came in weekends to carry out in-depth sequencing of Astrea and her parents,” says Ashley. The company’s scientists established that the gene variant was real and present in Astrea’s cells but not in her parents’ cells. That showed that it was a new mutation.
To find out if the particular mutation they had identified could cause long QT syndrome, Priest and Ashley gave the sequence to a team of collaborators at Gilead, a company that designs drugs to treat the disease. Gilead reported that the gene variant would cause long QT syndrome. Moreover, they said, this particular mutation was especially deadly.
Next the Stanford team wanted to be sure that Astrea really was a mosaic individual; they needed to show that individual cells actually had different genomes, some carrying the deadly mutation, some healthy. Each cell’s genome would have to be individually mapped. “That field was founded by Stephen Quake,” says Ashley, “so having him here at Stanford, I called him and asked if he could help.” Quake, PhD, a professor of bioengineering and of applied physics, and his team were able to look at individual cells from Astrea’s blood and show that most of her cells had normal genes, but 8 percent carried the mutation for long QT syndrome. Astrea’s blood cells were definitely mosaic. But it wasn’t yet clear if her heart tissue was also a mosaic.
As the weeks passed, Astrea began to struggle. Despite the implanted pacemaker and defibrillator, she developed an enlarged heart. At 7 months old, she looked healthy to her parents. But the defibrillator was going off frequently and sending silent signals to the hospital staff. Alerted by the signals, Astrea’s doctors told Tsoi to bring her baby in; Astrea was in grave danger, they said.
Once at the hospital, her heart stopped again. She needed a heart transplant, and her name was quickly added to a waiting list for a donor heart. Tsoi says, “I thought it would be at least a year of waiting.”
But only five weeks later, someone from Stanford called. “Are you driving? Are you in a safe place?” the voice asked. It was news of a donor.
On the day of the transplant, Tsoi and Li took their two older daughters and picnicked on the Stanford campus, waiting patiently once more to hear news of Astrea. In the evening, when the surgery was over, they went to see her. “The first thing I saw was the monitor,” says Tsoi. “That was the first time I’d ever seen the green line — the heartbeat line — so stable and regular.”
As a transplant patient, Astrea’s problems weren’t over, but her healthy new heart enabled her parents and doctors to breathe easier.
The pressure on the genetics team to understand her condition was lessened. And the tissue from Astrea’s original malfunctioning heart allowed researchers to determine that, indeed, 8 percent of the heart cells carried the deadly mutation they believed had been causing her long QT syndrome.
By now, Astrea’s mosaicism had become a problem of intense academic interest, initially triggered by the desire to help the baby. The team wondered if a heart with just 8 percent mutant cells could really have caused Astrea’s severe long QT syndrome. That was so few cells. Wouldn’t the 92 percent healthy cells have prevented such severe effects?
The Stanford team contacted colleagues in the computational cardiology lab of Johns Hopkins professor of biomedical engineering Natalia Trayanova, PhD, experts in the computer modeling of cardiac electrical activity. Their eventual computer model of a heart with a mosaic of healthy and mutant cells in the organ’s electrical tissue acted exactly the way Astrea’s real heart did. “It was an important moment: A mosaic heart really could cause heart block and cardiac arrest,” says Ashley.
Might other heart patients be mosaics?
The team had solved the genetic mystery. Not only did they diagnose Astrea’s problem but they had also revealed a new way to identify what might be causing genetic diseases that have no obvious source.
“We’d thrown everything we had at diagnosing the baby,” says Ashley, “but still we wanted to know, how common is this?”
To find out how often mosaicism might explain undiagnosed arrhythmia, the team partnered with a genetic testing company with a database of arrhythmia cases. “We asked them, ‘How many cases of mosaicism have you seen when you looked at genes that cause arrhythmia?’ The answer was about 0.1 percent,” Priest says.
The team’s work may offer a new way to finally determine the cause of arrhythmias that previously had been a mystery, says Priest. About 30 percent of heart arrhythmia patients don’t have a genetic diagnosis. “Maybe,” he says, “there are additional mutations that are in the heart only. Genetic tests are nearly always done on blood or other easily acquired tissues. So it’s easy to imagine a mosaic gene variant that occurs only in the heart and doesn’t show up in the blood.” The same reasoning, he says, could apply to other parts of the body.
“And that really is a brand-new phenomenon,” Priest adds. Until recently, he says, no one has thought of looking for mosaic gene variants as the cause of these kinds of diseases.
“I think there’s enough evidence now to suggest that a large number of people have some level of mosaicism,” says Ashley. That said, the genetic mosaicisms are probably harmless in most people. This one just happened to cause severe effects, even if only in a few cells.
Whether Astrea might have other health effects from the mutation is unlikely but not impossible. Astrea’s mother says that when her daughter was first born, “we asked, ‘Is she safe now? Is she stable now?’ One of the doctors told me, ‘We don’t know. But even for your two older girls, there’s no guarantee they’ll be healthy tomorrow. So treat Astrea like a normal kid and make every day count.’”
Tsoi says this advice inspired her to become a different kind of mother. “I used to be very strict, but since that day whenever I want to do something with the kids, I just do it as soon as possible.” She advises other parents, “Don’t think too much about the future. Don’t wait. Just do it.”
In September, Astrea celebrated her third birthday. She does cartwheels with her older sisters and loves to listen to the music from Frozen, which always comforted her through her many weeks in the hospital. Riding home in the car with her mother recently, Astrea said, “I don’t want to go home. I want to play outside.”
“And so we did,” says her mother.