Study: Evaluation of some direct-to-consumer genetic testing reveals inaccuracies and misinterpretations
|At a glance||In-depth|
|Guidelines||Resources and references|
Direct-to-consumer (DTC) tests, as the name implies, are genetic tests that a consumer can request directly from a company without involving a doctor or healthcare professional. DTC genetic tests are available for looking at common traits, making health predictions (e.g. presence of particular mutations for cystic fibrosis or sickle cell disease) and for ancestry analysis. In all cases, confirmation of any results by a clinical laboratory is needed to ensure that information is correct before any medical conclusions are possible.
More recently DTC genetic tests for determination of the presence of the 3 Jewish founder mutations in BRCA1 and BRCA2 have been made available. Note, this study evaluated DTC genetic tests that were done before 23andMe's more recently approved DTC test for the 3 common Jewish founder mutations was available. The quality of the test for the 3 common Jewish founder mutations was evaluated and approved by the FDA.
This study highlights the need for clinical validation of DTC genetic testing to confirm results and to ensure appropriate interpretation of results.
This study is about:
A clinical laboratory's evaluation of direct-to-consumer (DTC) tests and associated raw data analysis.
The FDA allows direct-to-consumer genetic testing for ancestry analysis, raw data, and limited clinical testing (including the 3 common Jewish BRCA mutations). These genetic tests come with a clear warning that any potential clinical variants must be reconfirmed by testing in a clinical laboratory.
Researchers at Ambry Genetics, a private clinical testing laboratory found that many mutations reported by DTC testing were absent (false positive) or misinterpreted.
- 40% of mutations (17 of 43) identified by DTC or subsequent third-party analysis of raw data were false positives: testing new samples showed that the individuals involved did not carry these mutations.
- Among mutations present, 8 that were classified as having "increased risk” by DTC or third-party providers were known to be benign, common variants or of unknown significance.
We note that this study predated the recent FDA approval of DTC testing for 3 common Jewish mutations in BRCA1 and BRCA2 (see prior XRAYS review). DTC testing for the 3 common Jewish mutations has met FDA standards for accuracy and reliability; these tests were correct and were not the type of tests examined by this study. Some DTC genetic tests (e.g. ancestry analysis) provided consumers with whole genome raw data. This raw data includes DNA sequence from regions that may have poor sequence quality or insufficient information. These regions would not be appropriate to determine presence or absence of a mutation (and are therefore not intended for clinical use).
What does this mean for me?
If you have had or are considering DTC genetic testing, be aware that the information provided is not intended for clinical diagnosis or risk assessment. All potential mutations found via DTC testing need to be reconfirmed by a clinical diagnostic laboratory. This study highlights the high potential rate of false positives—DTC results that incorrectly stated that individuals carried a mutation when they did not. These erroneous results can be a technical limitation of the DTC tests and analysis of raw data. Tests to reconfirm the presence or absence of these mutations are needed before this information can be used for clinical decisions.
The misinterpretation of some mutations as "increased risk" rather than benign mutations, which frequently occur in the population and are not associated with disease risk, points to the need for adequate genetic counseling from a certified genetic counselor or other trained health care professional.
It is important to know that DTC testing is typically not comprehensive. It may include only some and not all genes for a disease, and it may not test for some mutations in all regions of individual genes. Clinical laboratory testing is more comprehensive for a given disease or gene tested. FORCE recommends consulting a genetics professional (such as a board certified genetic counselor, medical geneticist or other provider with advanced training in medical genetics) before and after being tested so that the appropriate test is performed and properly evaluated. If DTC testing is pursued, it is important to bring any test results to your health care provider so you can discuss what you have or have not had tested and to ensure a complete understanding of your genetic tests.
Share your thoughts on this XRAYS article by taking our brief survey
This article is relevant for:
People who are considering or have had direct-to-consumer testing
This article is also relevant for:
Breast cancer survivors
Women under 45
Women over 45
Men with breast cancer
Be part of XRAY:
- Is genetic testing warranted for me, given my personal and family history?
- If testing is appropriate, what is the most appropriate and useful genetic test for me given my own medical history and my family history?
- How do I get confirmation testing for mutations found by a DTC test?
- How can I get a referral to a certified genetic counselor to discuss my genetic test results?
- Are the mutations found in my DTC test likely benign, pathogenic or of unknown significance? What follow-up is recommended?
Who covered this study?
What consumer DNA data can and can’t tell you about your risk for certain diseases This article rates 5.0 out of 5 stars
Healthcare providers can't afford to ignore direct-to-consumer genetic testing This article rates 4.5 out of 5 stars
Another reminder that consumer DNA tests are not 100% accurate This article rates 4.0 out of 5 stars
Consumer DNA tests are wrong 40 percent of the time This article rates 3.0 out of 5 stars
Nearly half of at-home DNA test results could be wrong This article rates 2.0 out of 5 stars
IN-DEPTH REVIEW OF RESEARCH
The FDA allows direct-to-consumer genetic testing for an individual’s ancestry analysis and non-disease status. These genetic tests come with clear warning that any potential clinical variants identified must be reconfirmed by testing in a clinical laboratory. However, people who have DTC testing to look at ancestry or other non-clinical purposes can obtain their raw genetic sequence. The caveat is that raw genetic sequenced may be unprocessed, without quality validation for many reported regions. There are third party companies that will interpret this information for a fee. However, the quality of DNA sequence or the method of DNA testing (such as single nucleotide polymorphisms or SNP testing) may not be suitable for analysis. This can leave consumers with the false impression that the information is accurate and known.
Researchers of this study wanted to know:
How DTC results compared to confirmation test results from their clinical laboratory. In particular, they wanted to determine the percentage of DTC results that were confirmed, and whether the variant classification matched that of the DTC testing company or third-party service.
Populations looked at in this study:
Researchers examined 49 cases that had been submitted for testing; each case was retrospectively identified as having prior DTC testing. The majority of these cases were from females (92%), unaffected with disease at time of testing (74%), 30-49 years old (53%), and Caucasian (51%).
This study from Genetics in Medicine in March 2018 involved 49 cases of DTC genetic tests that were tested at Ambry Genetics, a private clinical testing laboratory, to confirm or refute DTC results. Researchers found many mutations reported by DTC testing that were absent on Ambry's test (false positive) or misinterpreted.
- 40% of mutations identified (N=43) by DTC were false positives; testing new samples showed that the individuals did not carry the reported mutations.
- 26 of 43 mutations (60%) were confirmed to be correctly identified, including 2 in CHEK2, 13 in BRCA1 or BRCA2, including the 3 founder mutations and 4 unique mutations.
- 17 of 43 mutations (40%) were confirmed to be incorrectly identified, including 3 in BRCA1, 5 in BRCA2, 2 in CHEK2, 1 in TP3, 1 in ATM and 2 in MLH1.
- 8 mutations that DTC or third-party providers classified as having "increased risk" were mutations that are known to be benign, common variants or of unknown significance.
- These included 1 mutation each in ATM, BRCA1 and BRCA2 that experts classify as benign and occur frequently in the general population.
This study was limited to DTC cases referred to Ambry Genetics for clinical testing. In these cases, the consumer/patient and health care provider had chosen to confirm DTC testing, which may introduce some bias into the sample selection. The study authors acknowledge that the sample size is small (N=49), and they are collecting a larger series of cases to confirm these results. Confirmation by other clinical laboratories is needed to ensure that these results are generalizable. A further limitation is that only genetic testing for genes that Ambry currently evaluates were tested; only a subset of genes was evaluated.
An additional limitation is that this study does not evaluate false negatives in DTC testing - patients who received a result indicating that they had no mutations in particular genes who actually have a mutation present. Individuals whose results indicate no mutations that increase risk are unlikely to follow-up with clinical laboratory testing. Researchers did not report evaluating any cases of this type.
Because DTC testing is not designed to be used for clinical purposes, all DTC tests need to be confirmed by clinical laboratory testing. The presence of a DTC-identified mutation must be confirmed to rule out a false positive result. Misinterpretation of the clinical significance of particular mutations by DTC companies or third-party entities has occurred. Consumers are advised to consult a genetic counselor or informed health care provider to determine the most appropriate test given their situation and to evaluate their genetic test results. DTC testing may (often) not be comprehensive; absence of mutations in a DTC test may not rule out the presence of a mutation in all genes or all gene regions associated with hereditary breast cancer or other hereditary diseases.