The good news first: It is an exciting time in genetics. With recent technological advances and genetic discoveries, the clinical community is really just starting to explore personalized genetic medicine.
“Just recently, in December 2019, the ClinVar database celebrated its 1 millionth submission to the database, from 1,300 different organizations in over 70 countries, which goes to show just how common genetic testing has become globally,” said Laboratory Director Jennifer Schleit from Blueprint Genetics.
But the incredible growth in genetics has left us with some misconceptions about testing.
“There are myths about genetic testing I hear almost daily working with rare disease clinical interpretation, as well as during scientific conferences I attend,” Schleit noted.
Myth #1. Copy number variants (CNV) cannot be detected by next-generation sequencing
CNVs are becoming increasingly recognized as an important cause of many genetic diseases. Traditionally, CNVs have been detected for example by karyotype analysis, fluorescence in situ hybridization (FISH), MLPA or microarray. With new advancements, it is possible to detect a wide range of CNVs, even the smallest events, with next-generation sequencing (NGS) technology.
“CNV analysis performed in parallel with sequence analysis is a powerful diagnostic tool, not to mention very time-efficient,” Schleit said.
What this actually means for patients is that if CNVs are not included in the analysis, the molecular diagnosis might be missed. Also, it is not uncommon to come across cases where previous chromosome microarray testing was normal, but the diagnosis is due to a CNV detected with NGS.
“CNVs are an important disease mechanism that should be assessed in all patients with a suspected inherited disorder. The reliable calling of CNVs, especially smaller ones, requires a multifaceted approach including uniform sequencing coverage, careful review for called CNV events by laboratory geneticists to identify likely false-positive findings, and confirmation of those findings with alternative methods,” Schleit emphasized.
Myth #2. Diagnostic yield is a good indicator for genetic test quality
By definition, diagnostic yield means the proportion of patients in whom a medical technique yields a definitive diagnosis out of the total number of patients who receive the diagnostic procedure. In research, diagnostic yield is a very common metric for reporting results and assessing testing.
“Just by browsing the abstracts from two big scientific meetings this past year, I found 50-60 abstracts from each that report on diagnostic yields. While in many ways this is informative and interesting, it should not be confused as a synonym for quality,” Schleit said.
For one, diagnostic yield is not a good indicator of test capabilities, nor does it specify if there have been different patient populations between reporting groups.
“This metric is very dependent on the patients referred for diagnostic testing. Also, test design and quality vary between labs, which can also affect the diagnostic yield. Are the differences in diagnostic yield due to the actual test performance or because one group just happened to have more patients who tested positive? Since labs are reporting diagnostic yield from different patient populations, there just isn’t any way to know. So, what you have to look at are the properties of the test itself.”
Schleit recommends browsing the test performance metrics instead to determine the quality of the test. “There is still a big need in the field for transparency in publishing quality metrices and limitations related to a specific test.”
Myth #3. Genetic tests have limitations
If I send a patient for genetic testing, does the test detect everything?
“There are number of limitations related to all genetic tests. The question is, how transparently are they stated in the test report or on the website that describes the test? There are disease and category-specific limitations in all medical testing that physicians should be aware of for informed decision-making.”
Schleit gives an example.
“In certain type of ataxias, the disease can be caused by repeat expansions, such as in spinocerebellar ataxias and Friedreich’s ataxia. That is why on our website, we state that the Ataxia Panel should not be used for the detection of repeat expansions. Is an NGS panel still a good test? Yes! A good test knows its limitations.”
Myth #4. Clinical information is not necessary for genetic testing
“Clinical information is actually one factor that can influence the success of genetic diagnostics. Often, there are a large number of variants detected during testing. Laboratory geneticists need to identify those that are actually relevant for your patient. You do not want a list of 50 variants. Having clinical information helps us narrow it down to the clinically relevant variants,” noted Schleit. “These facts are becoming better understood in the field.”
What is a lesser-known fact is that a good phenotype match may allow the lab to go the extra mile to establish a molecular diagnosis.
“Knowing there is a good phenotype match allows the laboratory geneticists and bioinformatics team to look more closely at the data for abnormalities that might otherwise be missed. If something is identified, we can then work with our Clinical Development team to design customized assays to confirm these more challenging findings,” said Schleit.
“As a laboratory geneticist, finding that diagnosis for the patient is the key motivation for me. Help me help you by providing good clinical information from the get-go,” Schleit concluded.