Direct-to-consumer genetic testing (DtCGT) – much to be concerned about
THE BUSINESS OF MEDICINE
By Norbert Gleicher, MD, Medical Director and Chief Scientist, at The Center for Human Reproduction (CHR) in New York City. He can be contacted though The Reproductive Times or directly at either ngleicher@thechr.com or ngleicher@rockefeller.edu
Following the bankruptcy and closure of 23andMe, there is a growing perception that direct-to-consumer genetic testing (DtCGT) has reached its peak in popularity. However, the author of this article disagrees and, by presenting examples from the field of reproductive medicine, offers evidence to the contrary. Preimplantation genetic testing for aneuploidy (PGT-A) and polygenic embryo testing (PGT-P), when used in conjunction with IVF, should be considered forms of DtCGT, as patients in most IVF clinics are the ones who decide whether their embryos undergo PGT-A and/or PGT-P.
Neither of these tests has demonstrated clinical utility in the context of IVF, and neither has received FDA approval for any application in assisted reproduction. On the contrary, both tests have become increasingly controversial. In fact, PGT-P has been declared "unethical" by several professional organizations on both sides of the Atlantic, based on current scientific understanding. A similar position from the American Society for Reproductive Medicine (ASRM) is expected soon, following a joint guidance document released by ASRM and the Society for Assisted Reproductive Technology (SART) at the end of 2024, which found no clinical benefit for PGT-A in the general population of women undergoing IVF.
With 23andMe now in bankruptcy, some believe that the golden days of direct-to-consumer genetic testing (DtCGT) are behind us—but we strongly disagree. On the contrary, we believe this sector of the medical testing market is only beginning to realize its full potential.
23andMe, in many ways, serves as a cautionary tale rather than a representative example of the field. While initially enormously successful—reaching a public market valuation of over $1 billion—interest in its test inevitably declined. After all, it's a one-time test, and from day one, the potential customer base began shrinking daily. Aware of this limitation, the company understood it needed to diversify its revenue streams to sustain growth. Unfortunately, this is where the leadership faltered: the CEO and primary shareholder redirected the company toward pharmaceutical development, betting that its vast genetic database would accelerate drug discovery.
Unsurprisingly, this strategy proved overly simplistic. It failed to account for the lengthy timelines and exorbitant costs associated with drug development. Had 23andMe pursued a more realistic business model—especially given the considerable initial capital it raised—it might still be a thriving enterprise today. DtCGT continues to capture the public’s imagination and demand remains strong.
Consider, for example, preimplantation genetic testing for aneuploidy (PGT-A) in IVF. In a sense, this too is a form of DtCGT. Ultimately, the patient decides whether to test her embryos, and commercial PGT-A labs and IVF clinics are well aware of this, often marketing the service aggressively—and not always with full transparency.
Within this broader DtCGT context, a recent systematic review is worth highlighting. The review identified 40 eligible studies published between 2009 and 2022 (1). It found that between 1% and 57% of consumers shared their DtCGT results with a healthcare professional, while 19% to 76% of healthcare providers reported interactions with patients who had shared such results. Family members were informed in 18% to 98% of cases.
At the same time, concerns over DtCGT have prompted calls for tighter regulation (2). This has been a topic of ongoing discussion in the BMJ since at least 2011 (3). Yet, despite its wide-reaching implications, PGT-A—a test used daily to determine which human embryos are transferred—remains almost entirely unregulated. Given the profound, lifelong consequences associated with these decisions, the FDA’s failure to regulate PGT-A is nothing short of shameful.
Returning to 23andMe: its disease risk predictions have long raised eyebrows among experts. Unless a disease is monogenic (i.e., caused by a single gene), its frequency depends on a complex interplay among numerous genes. This makes polygenic risk scoring a highly complicated and, at present, imperfect science.
By the time of its shutdown, 23andMe reportedly offered predictions for 240 health conditions, according to a recent Nature article. (4) Starting with Parkinson’s disease and coeliac disease—both of which received early FDA approval—the company expanded to 50 diseases. This at a time when clinical medicine is still grappling with the viability and utility of polygenic screening for even the most common conditions. Unsurprisingly, many clinicians have questioned the reliability of these reports.
And if, as discussed above, we categorize PGT-A as a type of DtCGT, what then of polygenic risk screening for human embryos in IVF? This service is already being offered by a handful of commercial labs and IVF clinics in the U.S., as detailed in a compelling three-part New York Times series on IVF by Anna Louie Sussman, who herself underwent the process (5).
Interestingly, although Sussman devotes nearly the entire second part of her series to polygenic risk screening of embryos, she fails to mention the still-growing use of PGT-A, which has become routine in over half of all U.S. IVF cycles. This despite a 2024 formal statement from the American Society for Reproductive Medicine (ASRM) and the Society for Assisted Reproductive Technology (SART) concluding that PGT-A has not improved outcomes in the general fertility population.
While Sussman thoughtfully addresses the ethical, moral, and technical issues surrounding polygenic risk scoring, she overlooks the fact that multiple major professional organizations—both within reproductive medicine and genetics—have formally advised against such testing in embryos, declaring it unethical based on current scientific knowledge. ASRM is rumored to be preparing a similar statement (better late than never), though it is doubtful this will deter the labs and clinics already offering the service. That level of enforcement can only come from the FDA.