Solving the VUS Problem: How recent research on genetic testing can help us report better results
As our understanding of genetics has grown, so too have our genetic testing panels. Take, for example, testing for hereditary breast cancer: a panel which once included only BRCA1 and BRCA2 now tends to include dozens of genes associated with breast and other cancers. In general, this growth is a good thing. It’s helped more individuals get the answers they and their care teams are searching for. But more genes also mean more variants of uncertain significance on patients’ reports, and whether that is a good thing is far less clear.
A variant of uncertain significance (VUS) is just what it sounds like: a change found within a patient’s genes that may or may not be harmful. We don’t have enough evidence to call it pathogenic, but neither do we feel confident calling it benign.
Originally, it made sense to report any and all VUS for a given test, because a VUS could be the cause of disease. At the time, that likelihood of causality was relatively high — not only because panels were small and focused on only a few genes (resulting in a higher chance of correlation), but also because of the types of patients being tested. Genetic testing was costlier and less common than it is today, so tests were generally limited to those whose case clearly involved a hereditary disorder and a strong indication for testing. This meant that the person being tested was particularly likely to have the disease being tested for, and therefore any VUS was more likely to be relevant.
Today, costs are lower, and our understanding of phenotypic spectrums is wider. Now that genetic panels are more all-encompassing and are ordered for more people, a VUS is less likely to be causal. And in fact, reporting a lot of VUS can in some cases result in more harm than good. When providers are uncertain as to the relevance of a VUS, the confusion can result in a wild goose chase for answers that don’t exist. Furthermore, when VUS are misinterpreted as causal, there arises the possibility of medical mismanagement — the last thing that any patient, provider, or geneticist wants.
This may seem like a bleak outlook, but we do have other options. When it comes to VUS reporting on panels today, we can find potential solutions by looking at other kinds of tests. In fact, two of our team members here at Fulgent recently contributed to research showing that genomic testing yields fewer VUS than multi-gene panels (you can see the paper here). By looking at how and why this may be, we can find solutions for reducing VUS to the benefit of all involved.
We can start by defining the differences. A multi-gene panel is a targeted panel that tests genes associated with specific symptoms or a suspected condition, like breast cancer or hearing loss. On the other hand, genomic testing refers to full exome or genome sequencing, meaning the scope of genes analyzed is much broader, usually in an attempt to find a genetic cause for an unspecified disease or symptom(s).
Because exome/genome sequencing covers so many genes, the criteria for which (if any) VUS will be reported has always been stricter. For these types of tests, before reporting a VUS a curator must consider: 1) clinical correlation (clinical notes are more likely to be provided for exome/genome sequencing than for panels), 2) segregation analysis (an evaluation of the likelihood that a genetic change traveling through a family is causing disease in the family), and 3) strength of pathogenicity evidence. According to the researchers, these considerations result in fewer VUS reported for genomic testing than for panel testing, even though genomic testing covers more genes. To come to this conclusion, they looked at VUS rates for over 1.5 million sequencing tests from 19 clinical laboratories in North America.
So, how can we use this new information to reduce unhelpful VUS reporting and improve patient results? If appropriate, a provider could order an exome instead of a panel. However, the decision to go that route is often influenced by other factors such as phenotype, reason for testing, and even considerations regarding the patient’s health insurance. Panels are frequently the more reasonable option.
In that case, we’re left with one solution: order a panel, but treat it like an exome. There are several ways to do this. Consider what is involved in exome testing that has historically not been a part of panel testing — for example, duo/trio analysis. Testing one or two family members in addition to the proband is often performed for exomes and reduces the number of VUS reported by taking into consideration the genetic changes of close relatives. When feasible, providers can order duo/trio analysis on panels to improve results.
Physicians can also include more clinical notes and phenotypic data when ordering panels. This kind of information is often required for exomes, but if it became common practice to include it for panels as well, it could substantially reduce unhelpful VUS reporting. The more clinical data is known, the better our curators can determine the likelihood that a given VUS is relevant.
And lastly, in cases where VUS are almost certain to result more in confusion than satisfaction, patients and providers should be allowed to opt out of having them reported at all. Reporting only variants that are strictly pathogenic or likely pathogenic can help narrow the focus to those results with known significance.
At Fulgent we provide the option to opt out of VUS reporting, as well as the option to perform duo/trio analysis on genetic panels. We also offer the ability to create custom, flexible panels in order to target only the genes that are likely to matter most for each patient — resulting in cleaner, more focused reports.
We believe solutions like these are the future of genetic testing. And while the field continues to move forward at lightning speed, we see this study as a sign that sometimes it’s important to stand back and look at what rules and recommendations can be refined to serve our patients even better. In that light, we want to thank everyone who contributed to this recent research, both within and beyond our Fulgent family.
– Kathryn Francis
