Nobody Likes Uncertainty

Guest post by Marisa Vomvos, M.S.

On October 16th, 2018, the New York Times published an article about variants of uncertain significance in genetic test results [1]. This article made reference to a study by researchers at Myriad Genetics, published in the Journal of the American Medical Association (JAMA), involving reclassification of variants found in hereditary cancer testing in 1.45 million individuals over a 10-year period [2].

Genetic testing is not always a yes or no question

It’s easy to make the mistake of thinking genetic testing is going to be fairly clear-cut, good or bad results. After all, amazing advancements have been made in sequencing technology and genomic knowledge in a relatively short period of time, and it can feel like we finally have all the answers at our fingertips. However, the truth is there’s still so much we don’t know. We’ve gotten pretty good at finding mutations, but not always at interpreting them.

So, what kind of results can you get from genetic testing? Results are commonly returned using a 5-tier classification system outlined by the American College of Medical Genetics and Genomics (ACMG) and Association for Medical Pathology (AMP) [3]. From one end of the spectrum to the other, the tiers are: benign, likely benign, variant of uncertain significance, likely pathogenic, pathogenic. Yes or no answers may be possible when it’s known that a specific mutation falls on either the benign or pathogenic side of the spectrum with strong evidence to support it. For example, in testing for Huntington’s disease, if an individual has a certain number of three nucleotide (or “letter”) repeats in one copy of the HTT gene, there’s no doubt they’ll develop the disease. On the other hand, there’s plenty of harmless variation in everybody’s genomes too, that simply impact traits like taste or don’t really have any noticeable effect at all.

Some results are less clear-cut. For example, if genetic testing is pursued for something like family history of breast cancer, one might receive a positive result for a mutation that is well known to confer a significantly increased risk of cancer, but this is still not a “without a doubt” answer about whether or not you’ll develop breast cancer. Of course, it’s good reason to undergo more rigorous screening or consider risk-reducing surgeries, but there’s still a chance of never developing breast cancer too, which can complicate these decisions. Thus, there are mutations that confer risk, mutations known to be disease causing, variants known to be safe, and there’s one more thing: variants of uncertain significance.

What is a Variant of Uncertain Significance?

It’s a bit of a mouthful, but it can be abbreviated as VUS, the tier at the center of the spectrum. A VUS is what it sounds like: a variant in the DNA sequence whose pathogenicity or lack thereof is uncertain because there simply isn’t enough data on it. Maybe there are conflicting reports on this variant; maybe there are very few or no reports on this variant at all, and its frequency in the population isn’t well defined. Whether or not it’s disease-associated or harmless is currently unknown and difficult to decipher. When someone receives this sort of a result, something along the lines of “we found something but don’t know what it means,” it may cause some anxiety. The best approach to it would be “innocent until proven guilty,” but thinking about it that way might be easier said than done when you’re on the receiving end. Most of us are uncomfortable with uncertainty, especially when it comes to our health.

The issue with reclassification

In an ideal scenario, these VUSs would be reclassified once enough data is compiled on them to make a call one way or another, and the individual who underwent testing and received the result would be notified. If it’s found that more people have this variant than previously thought and don’t seem to develop any associated health problems, the VUS could be downgraded to likely benign or benign; whereas if studies were published showing that individuals with this variant developed similar symptoms and disorders, the VUS might get upgraded to likely pathogenic or pathogenic. A more specific example can be found in the supplemental material of the aforementioned Myriad Genetics study [2]. The researchers detail a specific mutation in the gene BRIP1 that was initially classified as pathogenic because it results in a premature stop codon – in other words, the cell machinery stops making the protein that the gene codes for before it’s actually supposed to. One would think that losing a chunk of the protein would affect its ability to function; however, the premature stop is introduced toward the end of the gene’s sequence, so not that much of the protein is lost, and it might not be in an area vital for protein function. This resulted in the variant’s reclassification from pathogenic to VUS. To say the least, classifying and reclassifying variants is complicated, and can involve computational software and algorithms, functional studies, control sequence databases, extensive literature review, and determining the threshold of evidence needed to make a call [3].

At the end of the day, it’s data that is needed to drive reclassification. Millions of people have already gotten their genomes sequenced, but not all of that data is shared, and that data isn’t necessarily representative of every population. Much of the reference banks are relevant for white individuals of European ancestry, but not individuals of other races and ethnic backgrounds. It’s possible a variant compared to the reference will come up as uncommon, but it could actually be fairly frequent in another population. Ancestry testing is actually based off the fact that specific marker patterns in DNA sequences vary based on ethnicity.

At Myriad, one of the most established genetic testing commercial laboratories, the lab does reassess and reclassify these VUSs, and more importantly, regularly sends updated reports of reclassification to ordering healthcare providers. The reassuring finding to come out of the Myriad study referenced in the New York Times article is that, among the approximately 25% of VUSs that were reclassified between 2006-2016, 91.2% of them were downgraded to benign or likely benign categories, whereas 8.5% were upgraded to pathogenic or likely pathogenic [1,2]. Moreover, reclassification of variants originally classified as benign/likely benign or pathogenic/likely pathogenic was rare (0.2% and 0.7%, respectively). Still, this means that about 75% of VUSs were not reclassified during this period, leaving many individuals stuck in the waiting game. Additionally, although the majority of VUSs were downgraded, some people did receive news that their VUS was upgraded. While this might seem like bad news, it can be a relief to get a more definitive answer and it’s important in being able to make more informed decisions about their healthcare, which is what reclassification is all about.

Reclassification can be life changing since it can affect family planning and risk management strategies, and so its timely delivery to patients should be of the utmost importance. Medical intervention is, understandably, not recommended to be taken in response to a VUS, such as risk-reducing surgeries in the case of a VUS in a cancer-associated gene [3]. So, if someone waits years for their variant to be reclassified without taking action, only for that variant to be upgraded to pathogenic, it could cost them precious time. Never finding out about a VUS being reclassified might cause one to live with the uncertainty with no end in sight.

What can be done to get VUSs reclassified more often and efficiently?

For one thing, genetic testing labs can work to develop and implement more effective systems for reclassification, like that of Myriad’s, and be sure to update healthcare providers. Currently, many testing labs lack a real organized system for reassessing the VUSs they’ve found, and may not reach out regarding reclassifications, but instead only provide new information when contacted directly by a doctor, genetic counselor, or patient.

Another movement in the field that is attempting to remedy the shortage of data on human genetic variation is “Free the Data,” an effort in support of openly sharing any and all data in the field. In the vein of freeing the data, the NIH has supported the creation of open-access databases like ClinVar, where researchers, expert panels, and labs can upload their findings and reports about specific variants, and anyone can view the information online for free [4]. This could allow and empower healthcare providers and even patients to try to keep an eye on VUSs themselves. One thing to make note of is that it can be challenging to have quality control when it comes to completely open-access databases since almost anyone can contribute, but it’s a start and something that will need to be continuously improved upon.

Additionally, there has been a push for research projects looking to sequence the genomes of large cohorts in an effort to better understand variation and what it means, such as the NIH’s “All of Us Research Program.” All of Us’ goal is to enroll one million or more participants to compile a ton of data on genome sequences, environmental and lifestyle factors, health status, and more [5]. Programs like this are also advocating for increased diversity of participant pools, which would help assemble more representative reference banks.

It’s a good sign that variants of uncertain significance are getting press, as it’s needed to bring awareness to the issues surrounding reclassification and to the range of results people can receive through genetic testing. These articles shouldn’t scare you or deter you from getting testing done if it’s something you’d benefit from, they should just shed light on the realities of the complexity of variant interpretation. As genetic testing becomes more and more popular, and direct-to-consumer companies market testing as though it’s entertainment, it’s important for people to be informed about the challenges involved in the reporting of genetic testing results.

For Patients

• It’s a good idea to take things into your own hands – call your doctor or testing company every so often if you’ve received genetic test results that include a VUS to ask about reclassification.
• Make sure your contact information is up to date with your doctor, so that if a genetic testing lab contacts them about reclassification they’re able to reach you and pass on the news.
• Talking to a genetic counselor before getting genetic testing done can help you understand potentially unexpected findings you may receive, and what your level of comfort with uncertainty is; they can also help choose a testing company that they know has a system in place for variant reclassification. You can check out our genetic counselors here!

For Healthcare Providers

• When choosing a genetic testing lab, take into consideration the threshold of evidence they require for classifications, as well as what sort of system they have in place for reclassifying variants and notifying you and/or your patient about possible future reclassifications.
• Referring your patients for pre- and post-test genetic counseling can help alleviate any anxiety that may come from receiving an uncertain or confusing result. A genetic counselor can also help you decide which labs to order testing from. Grey Genetics has many different support options available for practices large and small.

References & Further Reading:

[1] Kolata, G. (2018, October 16). The Results of Your Genetic Test Are Reassuring. But That Can Change. The New York Times. Retrieved from https://www.nytimes.com/2018/10/16/health/genetic-testing-mutations.html

[2] Mersch, J., Brown, N., Pirzadeh-Miller, S., Mundt, E., Cox, H. C., Brown, K., . . . Ross, T. (2018). Prevalence of Variant Reclassification Following Hereditary Cancer Genetic Testing. Jama,320(12), 1266. doi:10.1001/jama.2018.13152

https://www.ncbi.nlm.nih.gov/pubmed/30264118

[3]  Richards S, Aziz N, Bale S, et al; ACMG Laboratory Quality Assurance Committee. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-424. doi:10.1038/gim.2015.30

[4] Introduction – ClinVar – NCBI. (n.d.). Retrieved from https://www.ncbi.nlm.nih.gov/clinvar/intro/

[5] All of Us. (n.d.). Retrieved from https://allofus.nih.gov/