|It took six years of tests before Audrey Gevorkyan (left) and her husband, Hakop, learned the cause of their daughter Vianna’s puzzling condition. Using the latest in genetic diagnostic testing, doctors at UCLA determined that Vianna had a mutation in the SCN2A gene that previously had been seen in only one other child.|
By Dan Gordon
Photography by Ann Johansson
The Gevorkyan family’s nightmarish six-year odyssey began with a shiver.
Vianna Gevorkyan was 6 months old in the spring of 2007, when her mother noticed the recurring ever-slight shaking that turned out to be infantile spasms. Soon, the seizures became more intense, and no medication could control them. Eventually, Audrey Gevorkyan and her husband, Hakop, were told their daughter had intractable epilepsy.
The initial diagnosis was followed by a seemingly endless series of tests to determine the underlying cause of Vianna’s disorder. There was a slew of MRIs and CT scans, all negative. There were spinal taps and muscle biopsies. Genetic testing available at the time provided no clarity. Every new doctor the Gevorkyans saw had a new suspicion, and a new procedure. The years brought cross-country travel, misdiagnoses and substantial costs. Vianna was left with scars from the multiple muscle biopsies, and still there were no answers.“She was having developmental delays because she was seizing so much,” Audrey Gevorkyan says. “Vianna couldn’t walk, sit, reach or use her eyes. And yet, none of the tests found anything wrong.”
In March 2013, three months after she turned 6 years old, Vianna and her parents came to UCLA for a new kind of diagnostic test that they hoped would finally unravel the mystery. Whereas previous genetic tests were limited to a single gene, this one sampled 20,000 genes, covering the entire protein-coding region known as the exome. It is within the exome that most of the disease-causing mutations in the human genome can be found. “We were told not to have high expectations, because we had already done a million dollars of genetic testing, and everything was coming back negative,” recalls Audrey Gevorkyan. “But this was different.”
The test’s results made it clear why the conventional method of zeroing in on a suspected cause was never going to succeed. Vianna’s epilepsy was the result of a mutation in the SCN2A gene, previously reported in only one other child. Now, as new therapies become available, the Gevorkyans and their doctors will know which ones are likely to help their child. As more children with the same mutation are identified through the whole-exome test, their symptoms and responses to interventions can be correlated with those of Vianna. And perhaps best of all for the Gevorkyans, the uncertainty and endless testing are behind them.
BARELY MORE THAN A DECADE AGO, one of the most-ambitious scientific projects ever undertaken culminated with the publication in 2003 of the “reference human genome” — the genetic blueprint of a human being. Over a period of 13 years, and at a cost of $3 billion, an international consortium of scientists sequenced all of the genome’s 3.3-billion base pairs. Armed with the prototype code, researchers had a foundation to identify genetic variants associated with human diseases. Subsequent advances in technology set the stage for another milestone in 2007: publication of the first individual genome sequence. Still, the time and money spent on the endeavor — six months, $4 million — confined it to research settings.
But around that time, fundamentally new DNA-sequencing technology was introduced, and suddenly scientists could run the tests at a wildly accelerated clip. Because the first-generation technique — known as the Sanger method after its inventor, Nobel Prize-winner Frederick Sanger — was painstaking and plodding, researchers would typically sequence just 100-to-200 nucleotides, or a small fragment of a single gene. With so-called next-generation (also called massively parallel) DNA-sequencing technology, they could sequence 6-billion nucleotides in one run — the entire genome. And as with all new technologies, the cost started high but dropped precipitously, to the point where, for a patient like Vianna Gevorkyan, the entire protein-coding portion of the gene, and that of both parents, could be sequenced in a few weeks at a cost of about $4,500.
In 2011, the UCLA Clinical Genomics Center was established as one of three facilities in the world (the other two are at Baylor and Harvard) to put DNA sequencing to clinical use. For patients suspected of having a rare genetic disease that has eluded diagnosis through conventional means, a single test surveys the protein-coding regions of the genome, where the vast majority of disease-causing variants are believed to lie. Whereas prior efforts were limited to recognizing a pattern of clinical symptoms as the basis for determining the genetic culprit and analyzing one gene at a time, the new center simultaneously sifts through virtually all of the 37-million base pairs in 20,000 genes in pursuit of the single DNA change responsible for the patient’s disorder. Nearly half the time, the test yields a definitive diagnosis for patients whose cases have baffled other specialists.
"This,” says Kenneth Lange, PhD, chair of the Department of Human Genetics at the David Geffen School of Medicine at UCLA, “is the finest fruit of the DNA-sequencing era.”
“To be looking at all genes and all diseases at once is a quantum leap that I never expected to see,” adds Wayne Grody, MD (RES ’87, FEL ’86), PhD, director of the UCLA Clinical Genomics Center. “I thought it would be 50 years before we could do this at any reasonable cost, but the next-generation sequencing has come around so fast, and the price has dropped so rapidly, that it’s now practical.”
MORE THAN 3,000 MENDELIAN DISORDERS, diseases caused by a mutation in a single gene, have been cataloged. Some are well-recognized and understood: sickle-cell anemia, Huntington’s disease, cystic fibrosis. But many are extremely rare, to the point of occurring in one family, and “certainly, there are thousands left to be discovered,” says Dr. Grody. When a child presents with severe developmental or other issues that defy diagnosis, at some point a specialist will suspect it is one of these rare syndromes. But determining which one and identifying the genetic cause can be a daunting challenge.
It’s commonly referred to as the diagnostic odyssey — the long, emotionally and physically trying journey that pediatric patients and their parents go through in search of an answer for symptoms that persist. Like the Gevorkyans, many families travel extensively for consultations with top experts, enduring a battery of tests and unsuccessful therapies as they run up substantial costs. “The amount of testing can be staggeringly large,” says Stanley F. Nelson, MD, co-director of the UCLA Clinical Genomics Center. “Patients go from specialist to specialist, with each throwing a new idea into the mix — more imaging, maybe a biopsy, sequencing this or that gene, doing a biochemical assessment.”
Top: Dr. Stanley F. Nelson is
Bottom: A genomics center
Families ride a rollercoaster of emotions, as hopes are raised and then dashed with each new test. “The uncertainty is really difficult,” says Eric Vilain, (RES ’98, FEL ’99), PhD, chief of pediatric genetics and co-director of the UCLA Genomics Center. “You don’t understand what’s happening with your child, what additional symptoms to look for, or what the future holds — is it going to be a deadly disease, is it going to progress, is it going to get better? With each new specialist, you think you are finally going to get an answer, only to be disappointed.”
In such cases, physicians have had to make an educated guess at the genetic disorder, then request testing of a single gene or small group of genes suspected of being associated with the disease, with limited success rates. In many cases, this is due to conditions never previously described in the medical literature. The UCLA center breaks that barrier by sequencing about 95 percent of the 20,000 protein-coding genes with a single test. Known as clinical exome sequencing, the test looks for the disease-causing variant within the exome.
Applying state-of-the-art next-generation sequencing technology, the UCLA team generates more than 10-billion bases of raw sequence data per patient (ideally, also the patient’s mother and father), so that, on average, every protein-coding DNA base is sampled more than 100 times. Using the center’s informatics pipeline, variants from the reference human genome are identified. A typical individual’s exome contains more than 20,000 of them, almost all of which are benign.
To identify the potential disease-causing variants, the Clinical Genomics Center team applies a series of filters to these data, based on the patient’s family history and other relevant components of his or her condition. A search is then conducted for all genes reported in the medical literature and mutations known to cause any of the patient’s characteristics. The results are presented to UCLA’s Genomics Data Board — a multidisciplinary group of experts consisting of pathologists, molecular geneticists, molecular cytogeneticists, clinical geneticists, genetic counselors and informatics specialists — for analysis.
The yield — percentage of times the center can identify the mutation and make a definitive diagnosis — ranges from one-third to one-half, despite these patients being the most complex referred to UCLA after they have stumped multiple specialists. Often, the diagnosis comes years earlier than it would have been made otherwise, saving untold financial and emotional stress while potentially paving the way to more-appropriate care. (If a causal mutation can’t be conclusively identified, the sequencing data are stored for future reanalysis as new findings are published.)
“Exome sequencing has changed the face of clinical genetics,” says Dr. Vilain. “When I started 20 years ago, if we could diagnose 5 percent of our patients, we were happy. Now, we’re closing in on 50 percent with the most-difficult cases. Our turnaround time is much faster. And we can give families a much-clearer picture of what’s going on.”
UNFORTUNATELY, DIAGNOSIS OFTEN CARRIES WITH IT BLEAK NEWS. “Sometimes, we are giving people answers that are crystal clear and devastating,” says Dr. Nelson. “The information may predict an extremely harsh course for the child. But for such cases, most parents would rather know
than have to watch this unravel in front of them.”
“It provides a degree of closure,” says Julian Martinez-Agosto, MD (RES ’03, FEL ’05), PhD, a pediatric geneticist who conducts research on rare diseases and sees patients at the medical-genetics clinic. “The family can move forward, know what to expect and make plans.” Dr. Martinez-Agosto notes that the center’s involvement with the family doesn’t end with the diagnosis. “We participate with our colleagues who are specialists in other areas of medicine, coordinating multidisciplinary care that will address the complications associated with the genetic changes,” he says.
Beyond any catharsis, there are several important practical benefits to the diagnosis. For one, it can alert physicians to the possible presence of other conditions known to accompany the underlying mutation, some of which may be treatable. As targeted therapies emerge, the medical team can be on the lookout for those that address the patient’s disease. For some genetic diseases, there are existing treatments, and understanding how people with the same condition fared can provide guidance to the best approach or point them toward the ideal clinical trial. An example is muscular dystrophy, one of Dr. Nelson’s areas of expertise, for which a variety of molecular mechanisms have been described.
An early and accurate diagnosis can also provide critical information for family planning — informing couples about the risk of the same mutation appearing in any of their future children and enabling them to choose prenatal testing to avoid a recurrence. For years, the Gevorkyans delayed a decision on having more children out of a concern that the genetic cause responsible for Vianna’s condition would be repeated. Now they no longer need to wait: They know the risk is small and that they can test for the mutation before going forward with a pregnancy.
|Dr. Stanley F. Nelson discusses family pedigrees, which provide an important insight into the causal mutations, during
a group meeting of the UCLA Clinical
IT’S NO ACCIDENT THAT UCLA IS AT THE VANGUARD of bringing DNA sequencing to clinical fruition. “It takes a combination of skills to make this work, to be able to sift through this mountain of information and find the gold amid tons of rubble and rock,” says Dr. Lange. Few academic centers can boast both the breadth of expertise and the collaborative environment that exists on the UCLA campus. The Department of Human Genetics houses leaders in both clinical and laboratory genetics who have worked closely to apply the next-generation sequencing technology in new ways. The ambitious endeavor calls on the resources and expertise of the Department of Pathology and Laboratory Medicine, under which all clinical testing occurs at UCLA, as well as from fields ranging from pediatrics to bioinformatics.
That expertise — medical geneticists, molecular pathologists, genetic counselors and experts in next-generation sequencing and bioinformatics — is gathered in one room for weekly meetings of the Genomics Data Board, where the group reviews and interprets the individual genome sequences that emerge from the bioinformatics pipeline. The diversity of strengths is embodied in the leadership of the board and the center. Dr. Nelson and his lab have pioneered both sequencing capabilities and the bioinformatics needed to sort through the results. Dr. Grody is at the forefront of finding the clinical relevance in the complicated information coming out of the laboratory. And Dr. Vilain, as a medical geneticist who sees patients, brings the clinical expertise and the ability to straddle the laboratory and patient-care realms.
The remarkable nature of the technology and its implications are reflected in the wonderment with which Dr. Nelson — no stranger to dramatic advances in genetics and genomics over the course of his career — views the meetings of the Genomics Data Board. “Every week we review a series of cases of kids with very-difficult-to-diagnose genetic diseases, and in many of these cases, we uncover the exact mutation underlying an incredibly rare variant,” he says. “Each of these kids teaches us something profoundly important about how humans develop. That’s a contribution we ought to take seriously and use for the next generations.”
|Clinical exome sequencing is enhanced by the use of next-generation sequencing machines that can decode billions of bases every hour.|
Dr. Nelson explains that the molecular diagnosis will often affect the description of the rare disease, given that the patient may share only some components of the trait that was diagnosed or might have different symptoms from those previously associated with it. And in about 7 percent of the cases, the group diagnoses a gene mutation never previously associated with a human genetic disease. “Then we can do the laboratory work to confirm that, when mutated, the gene indeed causes the disorder,” Dr. Nelson says. “It’s a spectacular source of gene variants that can contribute to major differences in human development.”
The ability of UCLA’s exome-sequencing test to pick out a single base-pair change from the 3-billion base pairs in the genome has also forced the board to wrestle with a difficult ethical question: when and how to report incidental findings, those that have nothing to do with the diagnosis but could weigh on the future health of the patient, or that of his or her family members. In some cases, these are “actionable” — knowing of the mutation could spur treatment — but often the knowledge doesn’t change anything from a medical standpoint, other than alerting the patient to the potential for peril. Dr. Grody, who was president of the American College of Medical Genetics when the new technology took off, calls it “the toughest medical ethics issue I’ve dealt with in my career, and I’ve dealt with a lot of them.”
As an example, he offers the hypothetical case of a 3-year-old girl with hearing loss. In searching for a mutation in one of the deafness genes, the UCLA team finds a mutation in a familial cancer gene, such as BRCA-1 or -2. “That’s a test we would never normally do on a child, because it’s an adult-onset disease with no childhood intervention, yet now we’re stuck having seen this result,” Dr. Grody says. “You can argue that the family came in only for the hearing-loss issue — don’t burden them with this risk of breast cancer that’s not going to occur for 40 years, if ever. Others say, what if the child got the mutation from her mother, and her mother is at risk right now; don’t you have a duty to warn?”
The issue continues to be the subject of active debate. “This is a recurrent question that all of the patients ask me — ‘Are you going to tell me other things?’ — and our thinking is changing on this issue, as families have pushed for us to let them decide for themselves,” says Dr. Vilain. The American College of Medical Genetics has published a list of mutations associated with specific diseases — in most cases, familial cancers — that, because they are actionable, should be reported when found.
The Clinical Genomics Center provides extensive pre- and post-test genetic counseling, advising families of both the power and limitations of clinical exomesequencing — and of the possibility that while the process may yield answers, the news may not be hopeful.
|Dr. Hane Lee, assistant researcher in the Department of Pathology and Laboratory Medicine, leads the bioinformatic assessment of DNA sequences.|
“A lot of families are information-seekers and want this state-of-the-art test and to know the diagnosis no matter what,” says genetic counselor Michelle Fox, MS, LCGC. “Others decide they aren’t ready for it right now, preferring to come back only later when, for example, there is considerable developmental delay. The important thing is not to miss anything that can point to an effective treatment.”
Families are also advised about the potential for incidental findings and their implications — both medical and psychological — to ensure that they fully understand and are providing informed consent to receive the information. Some are interested only in the result related to the diagnosis, but others want to know more. For those families, genetic counselors explain that only certain types of incidental findings are reported — specifically, those that can be acted on through treatment or increased surveillance. The list remains in flux, and how to interpret and report incidental findings are questions tackled by the Genomics Data Board on a case-by-case basis.
“This is still a work in progress,” says Naghmeh Dorrani, MS, LCGC, genetic counselor for the UCLA Clinical Genomics Center. “But it is also a very exciting time. We are making diagnoses for some families we have been seeing for 10-to-15 years, and often they are for disorders so rare that we would never think to test patients for them.”
THE ETHICAL, SCIENTIFIC AND CLINICAL CHALLENGES — as well as the benefits — will surely grow exponentially in the near future, because such testing is likely to become more routine.
Not long after the reference human genome was sequenced in 2003, a grand challenge was issued within the field to push technology development to the point where an individual’s entire genome could be decoded for less than $1,000. What seemed unthinkable at the time is within sight. “We’re not there yet, but we keep moving in that direction, and some of the technology that makes it possible has already been described in the literature,” says Dr. Nelson.
“This kind of testing is going to be faster and cheaper, making it more accessible and broadening its applications beyond rare disorders,” adds Dr. Vilain. “Instead of doing the whole exome, we will be doing the whole genome. That will give us even more complicated information to interpret, but as we do the test on a large number of patients, we will learn more about what’s important in genomic regions that we currently know little about.”
By the time the Gevorkyans’ diagnostic odyssey reached UCLA’s Clinical Genomics Center, the FDA had finally approved a drug, rufinamide, that could better control Vianna’s seizures. Her strength and energy level increased. The diagnosis confirmed why the drug was effective. And it enabled the family to move beyond the nightmare of uncertainty.
“Now we can focus on improving Vianna’s health,” says Audrey Gevorkyan. “You get so tired of hearing, ‘There’s another test, let’s try that.’ We can finally look at our child and understand that there’s an answer.”
Dan Gordon is a regular contributor to U Magazine.