Skip to main content
Premium Trial:

Request an Annual Quote

23andMe Begins Reporting PGx Variants With Drug Info Under Recent FDA Clearance


NEW YORK – Direct-to-consumer genetic testing firm 23andMe has begun offering pharmacogenomic reports to customers that include specific drug metabolism information and don't require confirmatory testing before a physician can use the information to guide treatment decisions.

The new reports are authorized under a 510(k) clearance the firm received from the US Food and Drug Administration last August. They are not available for all genes the company tests but only for specific variants in CYP2C19 that are known to affect the metabolism of two drugs, citalopram, an antidepressant, and clopidogrel, a blood thinner.

Nevertheless, the change is significant for 23andMe, which has been authorized by the FDA to sell PGx test reports since 2018 but was hobbled by a requirement that the reports avoid describing any association between detected variants and specific drugs. The approval also mandated that doctors order confirmatory testing before using insights from the reports to guide patient treatment.

This created a somewhat paradoxical situation in which the company could give reports to patients listing their genotype, but for this information to have any utility, doctors would have to order additional laboratory tests which have themselves not gone through the FDA clearance or approval process.

Beginning last Wednesday, for the first time, certain 23andMe customers were able to access preemptive genotype information for these two commonly prescribed medications, and their physicians are now able to immediately implement it in their care.

According to 23andMe spokesperson Aushawna Collins, about 22 million people in the US are currently prescribed citalopram and another 20 million have prescriptions for clopidogrel.

The company has calculated that about 30 percent of its customers will receive a “likely less effective” result for clopidogrel, she said in an email, a number rising to 50 percent among East Asians and South Asians. Also, about 3 percent of customers, and 12 percent of East Asians and South Asians, would be expected to receive a result indicating an "increased chance of side effects" with citalopram.

The new reports are offered as part of 23andMe’s larger PGx offering, which includes analysis of three genes that impact a total of 11 medications and is provided through the "23andMe+" product, currently priced at $169 for an initial test kit and $29 per year for an ongoing membership.

Studies have demonstrated that individuals with specific CYP2C19 variants may not respond well to citalopram and clopidogrel and could have a higher risk for serious side effects. These include a recent meta-analysis published in JACC Interventions, which showed that patients with loss-of-function variants in the CYP2C19 gene had a 30 percent reduction in ischemic events, including heart attack and stroke, when treated with alternative antiplatelet medications versus clopidogrel.

On the antidepressant front, 23andMe cited a 2014 JAMA Psychiatry study, which found that an estimated 12,000 patients in the US present to the emergency department each year due to selective serotonin reuptake inhibitor-induced adverse events.

Meanwhile, the firm's own research has found that "the majority of people have at least one genetic variant in the CYP2C19 gene that impacts how the body processes certain medications, including citalopram and clopidogrel," Collins added.

But despite this evidence for both demonstrated and potential benefit, PGx testing is not widely prescribed at the moment.

Leaders in the field have conjectured that this is because the information has to be preemptive to be useful, available automatically when doctors are considering a prescription as opposed to something they need to order on a case-by-case basis.

"23andMe’s mission has always been to give people direct access to their genetic information, [and] these reports underscore that commitment," Collins said. "[Our] reports allow customers to learn how they may process certain medications in advance of taking prescriptions and experiencing these challenges."

"Rather than being prescribed a medication and experiencing challenges such as adverse side effects or lack of efficacy, consumers can proactively share these informative 23andMe results with their healthcare provider," she added.

23andMe's launch last week comes at a complicated moment for firms marketing clinical PGx tests, as the FDA has increased pressure in recent years on testing and software companies to stop selling PGx tests without its approval or clearance.

In October 2018, the agency issued a safety alert calling out unapproved PGx tests and software used to make clinical recommendations as a public health risk. This was followed by a warning letter to at least one lab and the subsequent removal of its PGx testing services. In the aftermath, other labs confirmed that FDA had also asked them to cease linking gene results to drugs in their PGx reports.

This resulted in a backlash — including a coalition formed to file a citizen petition enumerating the agency's "unlawful" enforcement actions — but also the genesis of new efforts by laboratory associations and consortia to create guidelines and best practices that might curb the overreach and overpromises that the FDA said prompted its crackdown.

By March of last year, the FDA had backed off to some extent, publishing a table of pharmacogenetic associations it deemed actionable, reassuring labs that it would continue to practice enforcement discretion and that it didn't view all PGx tests as problematic, and last fall joining a collaborative effort called STRIPE, which stands for Standardizing Laboratory Practices in Pharmacogenomics.

In practice, despite the FDA's increased activity in the space, various PGx companies have continued to report genotype-drug pairs or to link detected variants to their therapeutic implications. As such, 23andMe is far from the only source for individuals interested in preemptive PGx information.

Several competing offerings follow traditional lab testing models in which a physician orders a panel for their patient, while others, like Color, employ or contract their own physicians who order tests on behalf of customers.

Since the drugs that 23andMe's report concerns are prescription medications, customers need to interact with a doctor for the report to have an actual impact. The company operates a physician portal and makes it easy for patients to print out information on their results, but because physician interaction is still required, the practical advantage of DTC PGx testing over other models is limited.

For 23andMe, the provision of preemptive PGx results fits with its larger mandate and business model — to empower individuals with direct access to genetic information that might have relevance to their health.

Philip Empey is director of the University of Pittsburgh's Thermo Fisher Scientific Pharmacogenomics Center of Excellence. In that program, Pitt clinicians and researchers aim to return PGx results to what will eventually be up to 150,000 individuals, hoping to prove a positive impact of testing on healthcare utilization and spending, develop new standards for return of results, and inform best practices for patient education.

"I think [more than] 30 million people have already gone through consumer testing in the US, so [23andMe's share of that] is likely a sizable percentage of the US population [that could now access actionable information about their CYP2C19 status]," Empey said. "It's way more than [the number] tested in programs like ours," he added, "so if you're trying to disseminate or trying to increase the uptake of something that will have value to some patients, honestly, it's an exciting thing."

But simply increasing the number of individuals with access to PGx information isn't the end-all for this process, according to several clinicians who are not involved in large PGx programs like Pitt's.

Goutami Sanyal, a doctor at the University of Tennessee's Family Medicine Center, said that one barrier to patients benefiting from DTC PGx information is access. She noted that the low-income individuals who make up the majority of her patient community are unlikely to buy these genetic tests, and even if they did get access to such testing, they often lack even basic body literacy. As a result, they might not understand if, or how, they should share the results.

According to Janish Kothari, a cardiology fellow working in Pennsylvania's St. Luke's University Health Network, exposure to patients with direct-to-consumer PGx reports could increase awareness among his colleagues of PGx guidelines, like those maintained by the Clinical Pharmacogenomics Implementation Consortium, and the evolving literature to support them. These are currently not a prominent area of discussion or consideration at his institution, he added.

But to implement this in practice, there would need to be a clearer consensus around best practices, he noted, plus institutional support.

"If I'm putting a patient on Coumadin or warfarin, I know that other drugs that they're on could raise the CYP level. Certain patients are on over-the-counter remedies like St. John's wort and other things that they take for depression … that also have some impact on the CYP system," Kothari said.

"We take that into consideration. But those are all things that have been historically written in the literature and things that we've been educated on in medical school and on our board examination, that we're frequently tested on. So we have to actually get to a point where ... this [emerging] genetic information is ingrained into the medical literature in the same way."

Empey agreed, saying that despite the benefit of empowering consumers with this potentially beneficial information, there aren't necessarily good systems established for how those patients then move through the health system.

"I'm wearing the hat that this needs to be something that we do in more patients, and I'm really excited by it. But by the same token, there's still going to be a lot of limitations as to who has access, and how to actually use the results," he said.

This is the goal of large-scale preemptive PGx pilot programs he and others have implemented — which many hope will establish and codify decision support structures that can help support the wider distribution of this institutional-scale knowledge.

23andMe continues to stress that its FDA approval sets its reports apart from competitors in terms of proven validity. While other labs, following the FDA's 510(k) clearance for 23andMe, can now seek approval for the same CYP2C19 variants and drugs by establishing that their own tests are substantially equivalent, none appear to have done so.

Clinicians offered mixed takes on their perception of the value of FDA endorsement in this context. "When we are making decisions that are going to affect how we manage patients, we try to use evidence-based data and we do rely on the FDA," Kothari said.

"But just because the FDA approves a company to do testing … I don't necessarily trust that initially," Sanyal countered. "It's great to hear though, that university centers [like Pitt] are [exploring this at scale]. That's actually more compelling to me than an FDA [nod]."