For years, entrepreneurs have envisioned an industry based on direct-to-consumer genetic testing for the science-curious, the worried well, the fitbit wearers. They’ve watched the cost of DNA sequencing drop from prohibitively expensive to just a splurge, with dirt cheap lurking on the horizon. They’ve seen online ancestry testing create a small army of amateur genetic genealogists. But health-related tests, like those that are intended to quantify your risk of heart disease, cancer, or diabetes, have gotten pushback from the FDA. Repeatedly, the agency has suggested that this sort of testing will not be allowed without requiring a consultation with a doctor, genetic counselor, or other medical professional.
Now, it looks like that’s about to change. On November 6, FDA Commissioner Scott Gottlieb announced a new regulatory system intended to reduce the obstacles for companies that want to market tests that make health claims related to genes. Instead of requiring a review of every test, the FDA will do a one-time assessment of each company that wants to offer genetic health risk (GHR) testing. If the company establishes itself as a trusted provider, it can then introduce new GHR tests without pre-market review — reducing time, expense, and other barriers to entry. Explicitly, Gottlieb stressed that the new rules are designed to optimize “an efficient pathway to bring these tests to consumers.”
While this announcement definitely rings the dinner bell on a whole Pandora’s buffet of direct-to-consumer genetic testing, it doesn’t mean everything will be on the table. The review of manufacturers should weed out some of the straight-up snake oil. More subtly, the category of GHR tests attempts to draw a line between tests that indicate risk and tests that are diagnostic. For example, under these rules, you may see genes for breast cancer susceptibility — like BRCA 1 and 2 — included in GHR tests, but you’re not going to see GHR tests that include VHL, because finding a mutation in the VHL gene results in a diagnosis of Von Hippel-Lindau, a genetic syndrome associated with a range of symptoms, including a high risk of cancer.
Diagnostic tests will still be considered medical devices., and remain subject to a more stringent FDA review process. They will be available only if ordered by a medical professional. In other words, the FDA wants diagnostic testing to remain in the clinical realm.
Differentiating between diagnostic testing and susceptibility testing is very important in clinical genetics. Everyone is at risk (and, spoiler alert, we are all going to die of something). Making sure that doctors, patients, insurance companies, and everyone else understands that an increased risk is not the same thing as a diagnosis is a crucial part of genetic counseling. So in some ways this is a really good place to draw a line.
But, while it’s very important, the line between susceptibility and diagnosis is also extraordinarily hazy. For example, carrier screens for recessive disease are considered GHR tests, because they say something about your risk of having an affected child, even though carriers themselves are usually perfectly healthy. Sometimes though, carrier screening can reveal serious health effects for carriers of some conditions, such as Duchenne Muscular Dystrophy, Fragile X, or hemochromatosis. In other cases, having the gene blurs the line between what is a risk and what is a diagnosis. For example, testing positive for Lynch syndrome means you are at an increased risk for colon cancer. It’s not diagnostic, but a patient with polyps who has Lynch syndrome is treated differently than a patient with polyps who does not. Is it still only a risk if it alters medical care? This question has not been answered in the clinic or the courts.
Still, as I said, geneticists and genetic counselors do differentiate between predictive and diagnostic testing. And it’s important to note that our experience is that explaining predictive information is often the more complicated of the two. Probabilistic information can be very hard to contextualize. For instance, many counselors will not order a test that identifies people at higher risk of Alzheimer’s disease, because they feel there are a lot of issues associated with that sort of emotional anguish and no compensating clinical value, since we have no treatment to offer. This may or may not be the right approach, but the debate illustrates how tricky it can be to prognosticate.
Another reason genetic clinicians are wary of GHR testing is that such predictive tests are generally far more likely to be wrong. We are getting very good at finding a genetic explanation for a condition, when one exists. But using genes to predict the future remains fraught. Are you really 30 percent more likely than your neighbor to get cardiovascular disease? The actual answer to that depends on genetics, fitness, diet, stress, previous life events, and luck — don’t ever underestimate luck. I’m not saying that breast cancer susceptibility testing won’t save your life, because it might. I’m just saying that as a rule of thumb, when someone takes out their crystal ball, you should take out your grain of salt.
For these reasons, the type of testing the new FDA regulatory structure will help bring to the market is mainly the type of testing not done in clinical settings. It’s ironic that in many cases that will mean outsourcing tests that present the biggest genetic counseling challenge to settings with no counseling at all. This is something that customers should be asked to think about before they click “order,” and responsible vendors will do that.
Health-curious consumers contemplating GHR testing should be aware that the availability of a direct-to-consumer test doesn’t promise certainty, it promises uncertainty. After all, if they knew for sure, they wouldn’t be allowed to tell you.