NEW YORK (GenomeWeb) – A new assessment of the clinical utility of exome sequencing suggests that the test can significantly impact clinical management. A study of 78 patients who underwent exome testing at the pediatric neurogenetics clinic of the Kennedy Krieger Institute in Baltimore showed that all patients and their families with a positive result – about 40 percent – saw changes in the management of their disease.
For most families, a molecular diagnosis enabled more informed reproductive planning. In addition, in a fraction of patients, the results triggered treatment with new drugs or discontinuation of existing drug treatments, enrollment in clinical trials of new drugs, additional diagnostic tests of potentially affected organs, increased disease monitoring, or a change in prognosis.
The results, published online in the Annals of Neurology this month, confirmed that the test, which the clinic has been offering to patients since late 2011, is clinically useful and has a high diagnostic yield. "Our objective of doing this test is not simply that we want to know what's going on, but also because we could potentially identify a disease that's treatable," said Ali Fatemi, director of the neurogenetics program at the Kennedy Krieger Institute and the senior author of the study.
For the study, he and his colleagues analyzed results from 78 mostly pediatric patients with unexplained neurodevelopmental disorders, including developmental delay, intellectual disability, cerebral palsy, and autism spectrum disorder, who had undergone clinical exome sequencing between 2011 and 2014 after a variety of other tests had provided no diagnosis.
The majority of the exome tests were performed by Ambry Genetics and GeneDx, and a small number were performed by Baylor College of Medicine's Medical Genetics Laboratory. Even though the variant calling and analysis pipelines of Ambry and GeneDx differ, the researchers did not notice major differences in the results they reported, Fatemi said, noting that both labs tend to be conservative about calling novel variants as pathogenic.
He and his colleagues also recently started interpreting the exome variants in-house in order to study whether clinicians who have actually seen the patients interpret the variants differently from the testing labs, which could impact the diagnostic yield.
In their published study, the test yielded either pathogenic or likely pathogenic variants for 32 patients, or 41 percent, allowing their doctors to make a molecular diagnosis.
In 27 of these families, or 84 percent, the results enabled the patient's parents or siblings to gauge the risk of disease recurrence, helping them with reproductive planning.
Two patients, or 6 percent of the 32, were started on new drugs, mitochondrial co-factor supplements, based on their diagnosis. For another three patients, or 9 percent, the diagnosis revealed that they qualified for ongoing clinical trials of new therapeutics, IGF-1 in the case of two patients and deferiprone in the case of another.
"It's still a relatively small fraction of individuals who can now be treated, but if you have a 10 percent chance that your child can be treated for a disease that was untreatable before because you didn't know what it was, for a parent, it's a big thing," said Fatemi, who also holds an academic appointment at Johns Hopkins School of Medicine.
In six patients, or 18 percent, the diagnosis suggested that other organs might be involved in the disease that had not previously been looked at because the patients' primary symptoms were neurologic. Those patients underwent further testing, including liver function tests, electromyography and nerve conduction studies, an echocardiogram, endocrinological and immunological evaluations, and a sleep study. According to Fatemi, some of the problems that might occur in these other organs are treatable, for example, arrhythmias and endocrine dysfunction.
For another five patients, or 16 percent, the molecular diagnosis led to the discontinuation of medication they had been given on an experimental basis. Three patients, for example, stopped taking mitochondrial co-factor supplements because the test revealed they did not suffer from a mitochondrial disorder. Another patient had been treated with carbidopa-levodopa for dystonia, but the diagnosis no longer provided a rationale for that.
Overall, the study's diagnostic yield of 41 percent was high – other studies of patients who had received exome sequencing as a second-line test reported yields on the order of 25 percent. One reason might be the relatively small number of cases, Fatemi said, and another the fact that the pediatric neurogenetics clinic tends to see cases where a genetic cause is highly likely.
Whole-genome sequencing could potentially boost the diagnostic yield further, but Fatemi does not plan to use it clinically anytime soon. "Most known pathogenic mutations are still exome mutations," he said. "Once there is more data, and more experience, [whole-genome sequencing] will be something we will probably approach as well. With the millions of regulatory sites that have been identified, it's going to be some time until we really have a more clear approach to analyzing the data."
Currently, the clinic utilizes exome sequencing as a "last resort" test after patients have undergone other tests without receiving a diagnosis, such as neuroimaging, metabolic screening, chromosomal arrays, or Fragile X testing.
But given the high diagnostic yield of exome sequencing in their study, the researchers are wondering whether it could be more cost-effective as a first-line test. "My assumption is that most [patients] had seen several doctors and had several tests done, which in summary probably cost more than this test," he said, noting that a cost analysis still needs to be done.
Prices charged by laboratories for clinical exome testing vary widely, ranging from about $5,000 to $20,000, and continue to change, he said. Rates can be higher if a lab bills insurance companies directly, and lower for institutions who take on the reimbursement themselves. In addition, some labs have special rates for patients without health insurance.
"The biggest challenge right now is the reimbursement," Fatemi said. The American College of Medical Genetics and Genomics published guidelines for the use of clinical exome or genome sequencing in 2012, but most health insurance carriers have no formal guidelines yet. Doctors need to submit a letter of medical necessity for each patient, he said, and reimbursement is decided on a case-by-case basis.
But this might change as testing laboratories gather experience from analyzing thousands of samples and the clinical utility of exome testing emerges. "This is pretty much becoming a standard, and it's changing management, so we think that [insurance companies] should fund it," Fatemi said.