乌鸦传媒

AUSTIN, Texas -- Whole exome sequencing is now affordable enough to use in clinical practice to resolve difficult diagnostic cases, but benefit to patients and their families is often elusive, researchers said here.

Six different physician groups presented cases at the Child Neurology Society (CNS) meeting in which they used the procedure -- determining the exact DNA sequences of all protein-encoding regions in a patient's genome -- to identify the genetic roots of unusual neurological syndromes seen in their clinical practices.

But speakers at a CNS seminar warned that these investigations seldom produce entirely happy endings for the patients and their families.

More often than not, finding the culprit mutation does not lead immediately to an effective treatment, these experts said. Although insurance sometimes will pay for whole exome sequencing -- which, at between $5,000 and $10,000 per case, remains expensive -- some patients or their families may have to foot the bill themselves.

And a genetic defect may lie outside the exome, or the condition may have an occult environmental cause that doesn't involve the genome at all. These and other factors may doom the sequencing effort to failure.

It can even get worse than that -- not only may the sequencing fail to identify a cause for the patient's illness, it can reveal other unsuspected disease-associated mutations in the patient or in a family member that may or may not eventually cause illness, but that definitely cause anxiety and expense.

How It Works

David Adams, MD, PhD, of the in Bethesda, Md., explained that whole exome sequencing ideally is performed not only on the patient but also on parents and siblings when possible. Inheritance patterns are important in identifying novel disease-causing genes.

Typically, is undertaken after other testing methods, such as conventional lab assays and single nucleotide polymorphism (SNP) analyses, have failed to identify the cause of a patient's symptoms.

"Exomes are a fairly poor replacement for gene tests and panels for high-likelihood clinical hypotheses," he said.

But, Adams emphasized, it should precede targeted sequencing of suspect genes when those have relatively low likelihood of causing the illness.

Clinicians wanting to attempt whole genome sequencing should send blood samples (preferred over cheek swabs) for the patient and family members to a reference laboratory, along with as much phenotypic data on the individuals as possible.

Adams recommended that clinicians shop around, not just for price, but to determine what services the lab offers. For example, some will automatically conduct SNP analyses as well as the sequencing.

He also noted that the sequencing itself may be only the first step in an investigation. In a really difficult case, it may be necessary to create mouse models with a variant gene to determine whether it generates a phenotype similar to the patient's.

Just sequencing the exomes in a patient and two parents will cost around $10,000, he said, but "an extended workup for a complex case may cost $100,000 or more."

He also noted that exome sequencing is not perfect. In the end, identification of a candidate gene variant requires judgments on the part of medical geneticists. As a result, false negative and false positive results are not uncommon.

How Useful Is It?

Poster presentations at CNS reported successful diagnosis with whole exome sequencing in several dozen patients with a wide variety of clinical presentations. In most cases, the ultimate diagnoses received ad hoc names since they represented never-before-seen conditions, such as "MECP2-related disorder" after the gene found to be affected.

Some of these reports simply identified the mutations discovered as the effort's principal outcome. For others, the only clinical impact mentioned was that families were offered genetic counseling.

However, some patients did appear to benefit. , and colleagues at Children's Mercy Hospital in Kansas City, Mo., performed whole genome sequencing on a total of 56 children with mysterious neurological ailments and found a genetic cause for 36. Of those, nine had previously unknown mutations that could not have been identified in any other way.

For four of the nine, treatments were changed or initiated as a result of the investigations, Le Pichon and colleagues reported. For example, one patient found to have a mutated gene for aprataxin was started on a low-cholesterol, high-protein diet with coenzyme Q10 supplementation. Another with an MTOR gene defect was to have been started on rapamycin, but died before treatment began.

But benefits were indirect at best in the other five, such as provision of genetic counseling to the patients' families and, in one case, the investigation "ended a long testing odyssey."

In one of the seminar presentations, , of Cedars-Sinai Medical Center in Los Angeles, described a case in which whole exome sequencing revealed a previous diagnostic error.

It involved a 17-year-old girl who had developed progressive speech and gait abnormalities starting at age 7, with cognitive decline and leg spasticity appearing later. The precise pattern did not match any known syndrome but had features of 10 different disorders ranging from Niemann-Pick disease to Rett syndrome.

Pierson said that, after lab work failed to identify a clear cause, her doctors had ordered targeted analyses of a variety of individual genes, at a total cost of more than $15,000, which produced nothing useful.

Whole exome sequencing revealed a mutation in the beta-galactosidase gene, pointing to a diagnosis of . Pierson noted that this should have been picked up in the girl's previous lab tests. At age 12, her beta-galactosidase levels had been checked but came back normal.

After the gene mutation was discovered, the enzyme was assayed again and levels were found to be just 2% of normal. Pierson said that, in all likelihood, the earlier test was simply in error.

Another case in which sequencing led not only to a cause but to a treatment was described by , of Children's Hospital of Philadelphia, at the seminar.

A 6-year-old boy began showing epileptic seizures and myoclonic jerks shortly after birth. He subsequently had no cognitive or motor development and lived in a state of profound static encephalopathy with frequent seizures. No solid diagnosis could be made after conventional lab testing and SNP analyses.

Sequencing identified a mutation in an protein gene. Because the mutation was so novel that its phenotypic effect was unknown, the workup included functional testing, ultimately showing that it led to chronic hyperactivation of NMDA receptors.

That in turn suggested a treatment strategy, Marsh said. Memantine (Namenda) inhibits NMDA receptor activity, so it was tried in the boy, leading to a significant reduction in seizure frequency.

There Are Downsides

However, Marsh pointed out that the treatment was started too late to salvage the boy's overall cognitive and motor development, and it remains unclear whether it would have done so if the diagnosis had been made sooner.

Marsh also presented two other cases in which whole exome sequencing had even less optimal results.

The outcome in one was simply bad all around, Marsh said. It involved a boy brought to Marsh's institution at 18 months of age with a variety of anomalies including a single kidney and autoimmune neutropenia as well as neurological problems such as language delay and torticollis. At age 4, he was diagnosed with autism spectrum disorder and seizures began the following year.

Whole exome sequencing failed to identify any candidate genes to account for his symptoms, Marsh said. But, in the course of sequencing the boy's parents as part of the effort, it showed that the boy's mother carried genomic variants that could cause serious neurological and cardiac problems, including heart failure and sudden cardiac death.

Although the mother is currently symptom-free, and could well remain so deep into old age, she will now need "close follow-up" for the rest of her life.

The end result, in short, was that whole exome sequencing did not produce a diagnosis for the boy and raised the whole family's stress level, Marsh said.

The other case involved a girl with physical features and symptoms of but also others that didn't fit with that condition. Whole exome sequencing showed mutations in two genes that could be disease-causing -- one associated with Nicolaides-Baraitser and another of unknown clinical significance.

As a result, the girl ended with two diagnoses. Although two might be considered better than none, Marsh said the girl's parents were very unhappy because there is no treatment other than supportive care for either diagnosis. All that could be offered was genetic counseling for the parents. "It was a difficult discussion," Marsh said.

Comment