NEWS CORNER

You Have a Right to Your Data…. but Is It Really Your Data if It’s Inaccurate?

How wrong would your supposed data need to be before you would rather not look at it?

Many direct-to-consumer (DTC) genetic testing companies give customers the option of downloading their “raw data.” This is a file that includes results from genetic testing beyond what was included on the report provided by the testing company. There’s typically no additional charge for downloading the data. Uploading to a third party service, often for a small fee, can provide at least some level of rough interpretation of the data, based on information initially pulled from publicly accessible online databases. Depending on the third-party service, this information may undergo some level of review or annotation.

It’s free. Or free-ish when you take any third party interpretation services into account, so— why not download it and take a look?

A broken clock is right twice a day. How often are DTC genetic testing results right?

Anecdotally, among genetic counselors, it’s long been recognized that the raw data from DTC companies contain many inaccuracies. Many labs providing DTC testing are not accredited by CLIA, which provides oversight regarding analytical accuracy. These DTC tests are generally designed for recreational rather than medical purposes. And, to be fair to the DTC companies, “raw” seems to be meant to communicate that this is data that has not been validated.

A recently published article in Genetics in Medicine reported on the results from one clinical genetic testing laboratory doing follow-up testing for 49 individuals who had found mutations, or variants, in their raw data. In each case, these were genetic variants that were reported to prevent a gene from functioning normally and so expected to confer well-established increases in health risks. The best known of the genes tested were the BRCA1 and BRCA2 genes. Mutations in the BRCA genes confer significant increases in the risks for breast and ovarian cancer.

What did the clinical laboratory find?

40% of the results were not confirmed. This means that the clinical laboratory did not find the variant that the direct-to-consumer testing company + third-party vendor had reported.
60% of the results were confirmed. This means that the clinical laboratory did find the same variant that the direct-to-consumer testing company + third party vendor had reported. For eight of these confirmed variants, however, the clinical testing lab interpreted the variants as benign changes, or polymorphisms—not as genetic variants or mutations that would be associated with increased health risks.

Genetic testing has gotten cheaper, but quality control measures still involve a greater cost.

Next Generation Sequencing (NGS) has made genetic testing relatively fast and cheap—at scale. But without certain quality standards, careful review of the data, and sometimes additional testing by other methods to clarify ambiguous results—the testing is quick and dirty. The data that makes it onto the reports of most companies is likely reviewed in most cases. I think? Without an outside accrediting body providing oversight, all you really have to go on is the word of the laboratory. Or the experiences of a journalist who sent one kit to each company and then wrote up an article suggesting which one might be the best Christmas gift.

CLIA accreditation is a basic requirement of any U.S. laboratory performing clinical, or health-related testing. CLIA standards are part of federal regulation designed to ensure the analytical validity of testing. In the case of genetic testing, this means that a CLIA-accredited lab has certain quality standards to maximize the chances of a true positive result and minimize the chances of a false negative result.

CLIA does not go so far as to regulate the clinical validity of the test. Even among CLIA-accredited labs, the interpretation of analytically accurate genetic testing results may differ. Some variants in DNA are relatively straightforward to interpret; others, not so much. Most but not all clinical laboratories use publicly available databases to some extent in their interpretation of the data. More reputable labs, however, view any information entered into an open-access and minimally regulated database with some skepticism. They also rely heavily on the knowledge of scientific experts, critical review of any published studies to interpret any findings, and internal data.

When interpreting the results of testing done internally, a laboratory has all the relevant information information to work with. Different teams of experts may head up the sequencing and the interpretation of results, but the teams are able to communicate with one another. A single accredited laboratory stands behind both the analytical and interpretive aspects of the genetic test. A third party service, by contrast, must often work without knowing all of the technical specifications of the test or how and why results are labeled in a particular way.

Genetic testing done in clinical labs is usually more targeted—and more comprehensive.

False positives aside, it is important to keep in mind that these DTC testing companies are generally testing only for SNPs (single nucleotide polymorphisms). These are single letter changes (A, C, T, G) at specific areas of interest on specific genes. Clinical laboratories typically do comprehensive analysis of genes of clinical interest. It may seem like you’ve gotten back a lot of data—and you have! But it’s not that much data relative to all of your genetic information. Just a single gene is usually consists of more than 10,000 letters, or base pairs (A, C, T, G).

Family health history may not be as exciting as your DTC test results, but…

Family health history may be less exciting than your DTC test results—but it may also be a better starting point for thinking about your personal health risks, the likelihood that genetic testing could be helpful for you, and which clinical genetic test would be most helpful for you.

Many people are adopted and just don’t have information about their family health history. Others feel that they can’t ask questions about their family health history due to family dynamics. If you’re someone who is in touch with your family and you are able to ask them— do it!

For Patients:

If you download your raw data, keep in mind that it may not be accurate.
If your report includes a concerning variant, keep in mind that there are at least three possibilities:
- 1) the raw data could be wrong;
- 2) the raw data could be right but the interpretation by the third-party service could be wrong;
- 3) the raw data could be right and the interpretation by the third-party service could be right.

Tired yet? This could also be a reason to just skip the raw data

If your report does Not include a concerning variant, keep in mind that
- False negatives are also a possibility
- Other clinical testing may look more thoroughly at genes that are relevant to your specific personal and family medical history.

If you’re interested in genetic testing for health reasons, you will be much better served by testing done in a clinical laboratory. Genetic counseling can help you understand if there is genetic testing available that would inform your medical management. For many health concerns, there is just no genetic testing that provides a lot more insight than your personal and family health history.

Hereditary Cancer is one area of genetics in which clinical testing can be very valuable in guiding a personalized plan of care to reduce cancer risks. For basic background information on cancer genetics and hereditary cancer, check out this page.

If you’re wondering if you meet criteria for genetic testing for common hereditary cancers, you can take this quiz. If you choose to share your results with us, we’ll follow up with you!

For Physicians:

If a genetic test was not done in a CLIA-accredited laboratory, it can’t be relied on to be analytically accurate.
Even for test results from a CLIA lab, it’s important to pay attention to the testing specifications: there may be significant differences from one lab to the next.
If a genetic test was done in a CLIA-accredited laboratory, the interpretation of any findings may differ from one laboratory to another. CLIA does not provide oversight related to the clinical validity of genetic testing.
Test selection matters: a negative genetic test result when the testing performed was more narrow than was medically indicated can be falsely reassuring.