March 30, 2021Miriam Bergeret, MSc
As the cost of gene sequencing continues to decrease thanks to advances in next-generation sequencing (NGS), more companies have started to offer direct-to-consumer (DTC) genetic testing services. Simultaneously, an increasing number of people are electing to submit their DNA for testing—the MIT Technology Review has estimated that by 2021 more than 100 million people will have undergone DTC genetic testing. Of course, if someone told you that you could receive a report of whether any nefarious condition or disease lurks in your genes for the simple exchange of a few cheek cells, wouldn’t you do it? Unfortunately, when it comes to current DTC genetic testing methods, this is more fiction than fact. Before you give away the rights to your DNA, you need to understand the key differences between clinical and DTC genetic tests.
What is direct-to-consumer genetic testing?
Unlike clinical genetic testing, which is usually ordered by a health care professional to investigate a specific medical question, DTC genetic testing is an elective test marketed directly to consumers that can be ordered and completed by anyone in the general public without the involvement of a health care provider. DTC tests are usually completed by mailing in a sample of DNA, usually a swab or saliva sample containing cheek cells, then receiving the results in the form of an online report through a website. DTC genetic testing is also known as at-home genetic testing, consumer-initiated genetic testing, direct-access genetic testing, home DNA testing, as well as ancestry/genealogy testing when the service aims to answer questions about family history.
There are several companies that offer different types of non-diagnostic DTC genetic tests, including tests for ancestry, disease-causing genes (genetic variants), genetic health risk and carrier status, and pharmacogenetics, i.e., genetic differences that can affect a person’s response to medications. Results from DTC tests should be reviewed with a primary care provider before making any health or lifestyle decisions.
Though not diagnostic, DTC tests have the potential to act as relatively inexpensive, large-scale population health screening tools to find people who will be potentially affected by genetically-linked conditions but who may not have direct contact with a health care provider. “There are a lot of people slipping through the cracks,” says Dr. Matthew J. Ferber, the founder and former director of Mayo Clinic’s Clinical Genome Sequencing Laboratory and the founder and current director of Mayo Clinic GeneGuide.
But one of the risks is when consumers believe that DTC and clinical genetic testing are interchangeable, he says. “If you have a phenotype and you think you have a family history of a hereditary cancer, you should have traditional clinical-grade diagnostic genetic testing performed.”
Clinical vs direct-to-consumer genetic testing
The differences between DTC and clinical genetic testing include access, cost, and accuracy, though the number one difference is that only clinical genetic tests are diagnostic. That is because DTC tests are not regulated by the FDA—most DTC genetic tests are lab developed tests or LDTs, and currently, only one company in the US has received marketing authorization from the FDA for some of its DTC tests. Before submitting your DNA, you should understand the key differences and limitations between clinical and DTC genetic tests, including what the test actually looks for and how, so you can assess its quality and properly interpret the results.
The true cost of direct-to-consumer genetic testing
A clinical genetic test may cost $1,000 to $2,000 because it involves a full investigation into the genetic cause of a disease, which sometimes includes using more than one sequencing method and manual investigation of gene variants. To keep costs low, DTC testing companies use low cost, automated tests with automated analysis and reporting, which may cost in the realm of $100.
"When a DTC test is limited, the results can be misleading because of what is missing from the tests."
One low-cost automated approach is to use pre-planned single nucleotide polymorphism (SNP) arrays that contain a predetermined catalogue of SNPs. In terms of analysis, SNP arrays can only identify gene variants included on the array. “Those types of methodologies are extremely limited in the types of things that they will find because they will not find new mutations,” says Ferber about SNP arrays. “The problem with that is that for conditions like hereditary breast cancer, you'll only see them once within a family, and then you'll never see them again in the general population.”
In other cases, DTC tests use targeted gene panels that test for specific disease-causing genetic variants, but here too there are challenges—DTC tests may only include some of the relevant genes, regions of genes, or variants relevant to a disease.
“This is very important for the consumer to understand—What did I get and what did I not get?” says Ferber. When a DTC test is limited, the results can be misleading because of what is missing from the tests, potentially producing a false-negative result.
In contrast, clinical genetic tests that target a specific set of genes to diagnose certain diseases or medical conditions usually include complete coding regions, as well as any noncoding regions that are known to be relevant to the specific, or suspected, health condition. Here, focusing on a set number of genes increases the sensitivity of the test, minimizing the potential for false-negative results. Health care providers also use exome or whole genome sequencing to conduct broader investigations into the genetic cause of a disease.
Is direct-to-consumer genetic testing accurate?
Common genetic changes that are disease-causing or disease-associated are referred to as genetic variants. But variant calling is not a concrete science and there is no global all-encompassing gene variant database, leaving some room for interpretation depending on which variant library is used to analyze sequencing results.
Variant databases and ACMG/AMP guidelines
To reduce variability and standardize results, reports based on clinical genetic tests classify variants according to standards and guidelines established by the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) for interpreting sequence variants.
Genetic variants can be benign or likely benign, or classified as pathogenic, likely pathogenic, or labeled inconclusive, i.e., a variant of uncertain significance (VUS). Clinical genetic testing reports on pathogenic, likely pathogenic, and VUS variants according to ACMG/AMP guidelines. However, DTC test results usually do not include VUSs.
Other databases include the publicly available Genome Aggregation Database (gnomAD) and ClinVar databases, as well as ENIGMA, which focuses on clinically significant breast cancer genes such as BRCA1 and BRCA2.
Depending on which database a DTC genetic testing company uses in their analyses, there can be a discrepancy between their interpretation of variants compared to ACMG/AMP guidelines. This can result in false-negative or false-positive results, where a DTC genetic testing company or third-party service may classify a genetic variant as “increased risk” while the same variants would be considered benign via clinical genetic testing. One recent study found that, of 49 patient samples tested by a clinical laboratory to confirm the presence of genetic variants identified through DTC genetic testing, 40 percent of the variants reported across multiple genes were false-positives.
Moreover, some health conditions are only associated with certain genetic variants, where someone affected by the condition is more likely to carry a variant compared to a healthy person, but it isn’t enough to cause disease without other factors, such as environment. Gene variants only associated with disease are not always included in DTC tests, and even when they are, they may not be fully sequenced.
The variability of variants
“The problem is that they’re a work in progress. They’re not perfect. Variant classification is a science, yes, but it’s also an art, and it’s not always black and white.”
Variant classification itself can also vary, as members within the medical or academic community can disagree about the status of some genetic variants and their role in a disease. “The ACMG/AMP variant interpretation guidelines were a huge step in the right direction in terms of trying to standardize how variants are reviewed and classified,” says Dr. Linnea M. Baudhuin, laboratory director in the Division of Laboratory Genetics and Genomics and professor of laboratory medicine and pathology at the Mayo Clinic in Rochester, MN. “The problem is that they’re a work in progress. They’re not perfect. Variant classification is a science, yes, but it’s also an art, and it’s not always black and white.”
“The problem is that they’re a work in progress. They’re not perfect. Variant classification is a science, yes, but it’s also an art, and it’s not always black and white.”
Variant classification depends on many factors, she explains, including:
- how much you know about the gene,
- how much you know about the condition,
- what information you have on the patient,
- what information you have about the variant,
- and what databases are out there.
In other cases, many genes or gene variants, in addition to other factors, may contribute to a disease, so it may not be possible to calculate someone’s risk of disease without a thorough medical history.
This variability is why clinical genetic testing reports on VUSs. “Not only do we not shy away from [VUSs], we feel compelled to report all variants of undetermined significance so that, just in case future literature should prove them to be disease associated, we didn't withhold any information,” says Ferber.
Moreover, what one DTC company classifies as disease-causing or disease-associated variants, another may classify as benign, and that goes for third-party variant interpretation services, too—if you request your raw DTC testing data, you can upload it to a third-party variant interpretation service to get another interpretation. Thus, submitting DNA to multiple testing or interpretation services can produce different results.
Potential for more harm than good
The main motivation to undergo DTC genetic testing is usually its perceived health benefit, but whether DTC genetic testing produces tangible health benefits for the masses isn’t clear. In a 2017 study, researchers followed more than 1,000 people who underwent DTC genetic testing for six months. They found that those who received elevated cancer risk estimates weren’t more likely to change their diet or exercise or engage in cancer screening to reduce the risk of developing disease any more than those who weren’t found to be at risk. Thus, whether DTC testing has clear health benefits is uncertain, especially considering the potential risk of receiving misleading results.
Post-test residual risk
One of the biggest risks associated with both clinical and DTC genetic testing is a false sense of security or reassurance. For many DTC genetic testing services, this risk isn’t insignificant, as they may only cover a subset of the known genetic variants that contribute to human disease. A negative test result is never truly negative—there is always post-test residual risk says Ferber, i.e., the chance that someone may carry a disease-causing genetic variant that was not detected by a test.
For example, one company’s DTC test for hereditary colon cancer has been controversial despite receiving FDA clearance in 2019. This particular DTC test only includes two of the most common genetic variants associated with colorectal cancer, even though this type of cancer doesn’t usually have an underlying genetic cause. People who test negative for these two gene variants via DTC testing may believe that they are not at risk of colon cancer, but in fact, the test doesn’t include the genetic variants for Lynch syndrome—the most common type of hereditary colorectal cancer. Similarly, the same company’s test for Parkinson’s disease only includes one genetic variant of LRRK2 and one of GBA, both genes associated with an increased risk of the disease. In both cases, receiving negative results does not mean you aren’t at risk of developing those diseases, just that you are likely negative for the specific gene variants tested.
“Even when we have a negative result through our comprehensive [clinical] sequencing, it still doesn't mean it's negative,” says Baudhuin. “There’s still genes we don’t know about, and technical reasons why we may not be able to find certain disease-associated variants that are present. So, there’s always a risk that somebody could have a genetic disorder.”
Some tests are ethnicity-specific
When undergoing DTC genetic testing, it’s also important to understand that some tests are ethnicity-specific. For example, a common DTC test for BRCA1 and BRCA2, both involved in hereditary breast and ovarian cancer, only tests for three variants common to people of Ashkenazi Jewish descent while there are more than 2,000 reported mutations found in these genes.
"Though many databases have expanded and now include more information about the Black, Hispanic, and Asian populations, there is still very limited data available for Native American and Middle Eastern populations."
This means that, for the almost 8 billion people without an Ashkenazi Jewish background, the results of this test are almost certainly irrelevant. “Most genetic causes of breast cancer are due to very rare variants and you can't find those unless you do a comprehensive sequence analysis of the genes involved,” says Baudhuin. But without knowing these details, there is a risk that consumers will interpret a negative result to mean that they don’t have any BRCA mutations, which may or may not be true.
Another problem related to ethnicity is that most databases were created using data from people of European descent. Though many databases have expanded and now include more information about the Black, Hispanic, and Asian populations, there is still very limited data available for Native American and Middle Eastern populations.
“We don’t have a database for Native Americans,” says Baudhuin, “we actually don't have any way to determine if somebody has Native American ancestry.” So, it’s entirely possible for a person with Native American ancestry to take a DTC ancestry test and receive negative results.
“We all have common genetic variants and a lot of them are ethnically associated,” says Baudhuin, so the lack of diversity in genetic databases can directly impact the results: “At Mayo Clinic, we get a lot of genetic testing samples from the Middle East, and it's sometimes hard to understand the results because we have limited knowledge about variants in that population […] Whenever we see a variant, we can’t say it’s benign,” says Baudhuin, “so we report out many VUSs in that population.”
Research has shown that most people have at least one pharmacogenetic variant that would affect the choice or dosage of a prescription medication. These tests examine genes that encode enzymes involved in the metabolism of medications, but sometimes also include genes that encode drug targets, transporters, and human leukocyte antigen (HLA) molecules.
But like DTC genetic tests, DTC pharmacogenetic tests also pose a risk, the biggest being that people may stop taking a medication right away based on their results before speaking with their health care provider. While one study showed that fewer than 1 percent of people who underwent DTC testing changed their medication without consulting a health care provider, 1 percent is significant when placed on a global scale.
Similar to DTC genetic tests, pharmacogenetic tests are also limited in the number and variety of variants tested, leading to the same concerns around false-negative results. In late 2018, the FDA issued a safety communication to warn consumers, as well as health care providers, that the results from many pharmacogenetic tests that claim to predict how someone will respond to a medication are not reviewed by the FDA and are often not based on strong clinical or scientific evidence. Because of this, the FDA recommended that all results should be discussed with a health care professional and confirmed through clinical pharmacogenetic testing prior to making any changes to medications.
Privacy and security risks of direct-to-consumer genetic testing
Beyond direct health risks, DTC genetic or pharmacogenetic testing comes with other risks. In the US, the Genetic Information Nondiscrimination Act (GINA) protects people from discrimination based on personal genetic data by health insurance companies and employers.
But DTC genetic testing doesn’t fall under the Health Insurance Portability and Accountability Act (HIPAA). Once you agree to a DTC company’s terms of service, you’re essentially signing away your right to your DNA data, giving the company full control about how that information is used.
In late 2019, the Pentagon issued a warning to military personnel that unregulated DTC genetic tests were inaccurate and could expose an individual’s personal genetic information.
Before undergoing DTC testing, consumers should understand that without HIPAA protection, individual companies can set their own data storage and protection policies, as well as policies around selling and disclosing consumer genetic data and using it for research, all of which are policies that can change—there is no guarantee that your personal genetic data will remain confidential.
Moreover, because DTC testing is not diagnostic, DTC companies encourage people to discuss their DTC testing results with a health care provider. However, once someone discusses their DTC test results with their health care provider, it becomes part of a patient’s medical record, which insurers can request access to.
How can direct-to-consumer genetic testing benefit society?
Though there are major differences between clinical and DTC genetic testing, and any DTC test results should be confirmed through clinical testing and genetic counseling with a trained health care professional, DTC genetic testing does have the potential to benefit society.
“If we can offer a test that's good enough from a sensitivity and specificity perspective, and cheap enough that it's financially feasible to sequence every adult in the United States, I know for a fact that we are going to find people who otherwise would never have been diagnosed until later in life when they have a life-threatening event,” says Ferber. “I’m trying to make sure that we've got the right educational support system in place, so that when lay people access consumer genetics, we have a support system for them to access if they do end up being one to two percent of people with a meaningful and actionable result on the back end of that testing.”