The genome is full of secrets. And some of them can kill you: Tiny alterations in key genes can send a newborn’s body into a downward spiral as poisons build up, seizures wrack the brain, and organs go into failure.
When doctors see a patient whom they suspect of having such a disease, they call in clinical geneticists, who have a keen eye for symptoms that point to specific conditions. Even so, for newborns with a genetic disease, death may come before an answer.
For older children whose symptoms were not evident immediately, their families must often go through what is called a “diagnostic odyssey” as clinicians conduct test after test trying to determine the cause. It’s a phrase that encompasses all the pain, fear, and uncertainty of not knowing what you can do to help your child, whether they will live to adulthood or die next year, and even whether you should have another child, knowing there’s a chance they will suffer in the same way.
In the last few years, however, something remarkable has happened—so quickly that clinical geneticists’ voices are filled with awe when they talk about it. Nearly science-fictional advances in the speed and cost-effectiveness of genome sequencing now allow physicians to sequence large swaths of their patients’ genomes, or even, in a few select places, the entire genome. If traditional genetic tests, which usually check for a particular disease, are like pointing a telescope at a distant star, then this is like standing in a planetarium surrounded by the Milky Way, lost in a sea of brightness and information. Last fall, two papers in JAMA Pediatrics showed the effects of deploying these methods on a large scale, with exciting results. The rise of these techniques signals a shift in how genetic diseases are diagnosed and treated.
Race Against Time
To give a sense of the stakes here, perhaps one of the most immediate changes is for babies who, in the normal way of things, might never have lived long enough for their parents to get a diagnosis, not to mention a hope of saving them. One such newborn recently arrived at the Rady Children’s Hospital in San Diego. The baby shook with constant, uncontrollable seizures, even when treated with the strongest anti-epileptic drug available. Though the doctors upped the dose to the point where the infant had to go on a ventilator, the seizing went on. “The EEG readings are telling us that this baby, as we watch it, is frying its brain,” recalls Dr. Stephen Kingsmore, who is the president and CEO of the Rady Children’s Institute of Genomic Medicine, which works with the hospital. “And so we are counting the days until this baby will die. We are aware that progressive neurologic damage is happening as we watch.”
Traditional genetic tests may take months or weeks to come back, by which time this child would have been long gone. At the Rady Institute, however, a whole genome sequence can be turned around in an average of four days. The Rady family gave $120 million to help establish the Institute of Genomic Medicine that Kingsmore leads, and it’s widely acknowledged as a leader in these techniques. Within 72 hours of the child’s arrival, geneticists were looking at a whole genome sequence, checking their suspicions against the data. And there the problem was: The infant had a mutation in a gene for an enzyme that controls the breakdown of the amino acid lysine.
“It turned out that the seizures were treatable with injections of vitamin B6, arginine, and a modified diet” with low lysine, Kingsmore says. These treatments will be lifelong, but they protect the brain from further damage. Ten days after being admitted, the baby went home.
It’s enough to make your head spin, especially since it wasn’t that long ago that single mutation tests were a pretty exciting technology, says Dr. Seema Lalani, a clinical geneticist at Texas Children’s Hospital and Baylor College of Medicine who led one of the new JAMA Pediatrics studies. When she started her training in 1999, such assays were already a big part of a geneticist’s toolkit. Then in the early 2000s, the chromosomal microarray (CMA), which allowed geneticists to look at multiple regions of the genome, came on the scene. If the genome is a book, she explains, then “CMA was able to help us with finding if there was a page or a paragraph missing.”
Still, as progress on sequencing accelerated, a new world came into view for Lalani and other geneticists. 2009 was a key year. In August, a team led by Jay Shendure published a proof-of-concept paper in Nature demonstrating that even if you just sequenced the portion of the genome that makes proteins—the exome—you could identify the gene behind a genetic disorder. In November, they published another study where they put it into practice: They discovered the gene behind Miller syndrome, a disorder whose precise cause was unknown, using exome sequencing.
That changed everything. “That’s when we knew it had really hit,” remembers geneticist Dr. Leslie Biesecker of the NIH, who was present at the American Society of Human Genetics meeting in Honolulu that year, when the researchers first released their work to the community. “The entire field of inherited genetic disorders was going to be cracked open. Progress was going to massively accelerated.” This was like getting all the letters in all the words of a very significant part of the book.
Indeed, research boomed. Between 2010 and 2015, about 250 new genes a year were connected to human disease. The number of genetic disorders with a known cause skyrocketed, and plans to take the technology from the lab to the clinic got into gear. “It happened faster than any of us thought and paid off more richly than any of us thought,” Biesecker says. Within a couple of years, there were labs set up where doctors could send patient samples for clinical sequencing.
Still, these kinds of sequencing can’t help with everything. Children with autism, for instance, don’t have nearly the luck with whole-exome sequencing as some others, Lalani says, noting that only about 4–5% of them get useful information. The sequences also don’t just come off the presses with an answer written in red ink. Instead, it’s just a jumble of letters that geneticists have to pore over carefully, going through a list of genes that their experience has taught them may go along with certain symptoms. Even then, mistakes can be made. “There are certain genes in the genome that are very hard to interpret,” Kingsmore says. In those cases, assuming a simple cause-and-effect relationship may be a mistake.
All the same, in a surprising number of cases, clinicians can go to their patients and their families with a useful answer. In their JAMA Pediatrics study, Lalani and colleagues used whole-exome sequencing with nearly 300 infants suspected of genetic diseases that came through their hospital doors over the course of five years, starting in December 2011. These were children who, like the baby in San Diego, were critically ill. In 36% of cases, a genetic diagnosis came back.
For some, the diagnosis meant a different course of action than might have been taken without the information. Sometimes doctors could tell families definitively that there was nothing that could be done for the baby, beginning the process of helping them think through what would come next. For others, the genetic information meant a better chance at life, pointing toward treatments like heart transplants or stem cell transplants and away from medications and other therapies that had no bearing on the disorder. As well, in a surprisingly high number of cases, Lalani says, the doctors could give the parents of children who had died information about whether future children would be likely to have similar problems.
The number of cases in which sequencing gave useful information is similar to what Kingsmore and colleagues reported in a study two years ago, suggesting that this may be the yield doctors can expect when sequencing infants suspected of a genetic disease in an intensive care unit.
In the other recent JAMA Pediatrics study, an Australian team focused on older children seen by geneticists at a clinic at the Royal Children’s Hospital in Melbourne, the kind of children for whom the diagnostic odyssey had gone on and on—in this case an average of six years. They were suspected of having a genetic disease, but had not had any sequencing. With sequencing of the whole exome, 52% of them received a diagnosis. For about a quarter of those, it changed their treatment plan. It also changed plans for future tests. For instance, says Dr. Tiong Tan, a clinical geneticist at Victorian Clinical Genetics Services who coauthored the study, one child had been lined up to have a spinal tap and a brain MRI. But they were canceled after the sequencing came back with a clear answer. “You can avoid a whole bunch of tests by making a specific diagnosis,” Tan says.
That’s an important observation, because another focus of the study was cost. While sequencing prices have plummeted, these tests still cost thousands of dollars. Insurance companies may question whether such an expensive option is worth it. Part of the goal with this study, Tan says, was to see whether doing sequencing when patients first come into the clinic results in not only a less painful, frightening, and dangerous process—about 60% of the children had had to go under general anesthesia at some point for a diagnostic procedure—but also a cheaper one.
As it turns out, it does. According to the study, it cost $6,838 less to just sequence the children right away than to go through a lengthier set of tests and visits with other specialists. The researchers hope that this study will encourage the Australian government to include whole-exome sequencing in the list of treatments that the national health service will pay for and help colleagues in other countries make similar cases to insurance companies.
While sequencing is changing how patients and families get answers, it’s also provoked some interesting observations about these diseases. More often than expected, when the results come back, the geneticists realize the child has a disorder that they didn’t know could look like this. Years of training and clinical observation have trained their eyes to certain symptoms, textbook cases. But about 30% of the diagnoses the Melbourne team saw were surprising. Often, Tan says, “the exome sequencing shows us an unknown form of a known syndrome.”
Lalani and her colleagues experienced this too. “We saw this recurrently, again and again,” she says. “About 38% had atypical presentation of their disorders” or a form that hadn’t been recognized before. As doctors continue to explore the use and usefulness of these tests, they may learn not only what to tell their patients and their families, but more about how these diseases work, potentially benefiting patients who have yet to be born.