Updating the DTC Debate: Trial by Press Release, More FDA Letters, the Problem of Pleiotropy and New RUO Guidance
Later today I will join several colleagues here in Chicago, IL at the American Society of Clinical Oncology (ASCO) annual meeting for a panel discussion on Direct-to-Consumer Genetic Testing for Cancer: What Physicians Need to Know (pdf). (Daniel MacArthur and Misha Angrist will not be on the panel, although each joined us in authoring the pre-conference paper.)
This will, I believe, mark direct-to-consumer (DTC) genetic testing’s formal debut at ASCO. It should also serve as another reminder that, despite its relatively small numbers (both in terms of dollars and customers), DTC genetic testing continues to exert an outsized influence when it comes to conversations about the future of genomic medicine. This is particularly true when the discussion turns to appropriate policy and regulatory oversight.
In advance of ASCO, here are several items of interest from the past few weeks in DTC genetic testing.
“Trial by Press Release.” That’s exactly what Daniel MacArthur of Genetic Future termed this week’s press release from the European Society of Human Genetics (ESHG) highlighting two presentations on DTC genetic testing at the recent ESHG annual meeting.
The studies themselves have yet to be published, but preliminary findings were meant to provide new evidence that (a) the risk prediction models employed by personal genomics companies are neither perfect nor consistent and (b) clinical geneticists are skeptical of the value of DTC genetic tests.
Neither finding is surprising, although, as MacArthur points out, the first study – from Rachel Kalf of Erasmus University – will need to be published in full before its claims, including the claim that one DTC company (deCODEme) utilizes fundamentally flawed risk prediction models, can be fully vetted.
The desire to encourage greater transparency and consistency among the risk prediction models employed by DTC genetic testing companies has been expressed numerous times in the past (higher-profile examples include current NIH Director Francis Collins, genomic pioneer Craig Venter and colleagues and last summer’s GAO report) and repeatedly embraced by DTC companies themselves (see, e.g., here and here).
When it comes to debating the future of the DTC genetic testing industry there are plenty of areas of legitimate disagreement. For example, should different types of DTC tests (e.g., carrier screening vs. pharmacogenomic testing vs. genetic ancestry testing) which make different types of claims be regulated differently? And how, if at all, should the concept of utility, whether clinical or personal, be incorporated into a genetic test evaluation?
But the need to improve transparency and consistency in DTC risk prediction models already appears to be shared nearly universally among DTC stakeholders.
The second study included in the ESHG release, conducted by Heidi Howard of the University of Leuven, Belgium, includes the finding that 63% of a “representative sample of clinical geneticists” from across Europe “wanted to proscribe whole genome scans carried out by DTC companies.”
That statistic has bounced around the internet echo chamber the past few days, but, again, is it offering us anything we did not already know (or at least strongly suspect)? As MacArthur points out, asking the current cohort of genetic testing gatekeepers how they feel about being sidestepped by genetic testing companies seeking to engage directly with individuals feels slightly rhetorical:
While it is important that personal genomics companies consult with medical professionals in devising their risk prediction algorithms and interfaces, regulation of genetic testing should not be based on the views of the traditional gatekeepers of genetic information.
As both MacArthur and Razib Khan of Gene Expression note, the appropriate question to ask is not whether clinicians believe genetic tests are harmful in the hands of consumers, but whether genetic tests are actually harmful in the hands of consumers.
Thankfully, a growing list of researchers is asking that very question, including Bloss et al. (see also a recent back-and-forth in the NEJM), Kauffman et al. and the recently launched Impact of Personal Genomic Services (IPeG) study (some details here), about which my co-presenter Stacy Gray will be providing more details during today’s panel.
As the Food and Drug Administration (FDA) and others continue to wrestle with the difficult job of shaping DTC regulatory policy, the data generated by these and other studies will be critical in ensuring that any policy which eventually emerges responds to the demonstrated challenges of DTC genetic testing, and not merely to hypothesized harms.
The FDA Sends Three More DTC Letters. Speaking of the FDA, nearly a year to the day after it sent its first letter of concern to a DTC genetic testing company (Pathway Genomics), the agency sent out its latest batch of DTC letters. The recipients this time: Lumigenix, American International Biotechnology Services and Precision Quality DNA.
Much has transpired since that first FDA letter – including many more DTC letters, a Congressional hearing last summer and a closely watched advisory panel meeting earlier this spring – but for now, at least in public, the FDA is continuing to employ the same case-by-case approach to DTC genetic testing oversight it has utilized from the outset.
Not everyone is happy with the FDA’s current approach. One of the most recent companies to receive a letter, Precision Quality DNA, has devoted a section of its company website to the FDA’s handling of DTC regulation, and has published its strongly worded reply to the agency’s most recent letter.
While it would be inaccurate to characterize Precision Quality DNA’s response as representative of the views held throughout the DTC industry, several of the points the company raises – including the need for greater transparency surrounding the FDA’s review and oversight of DTC genetic testing products – seem likely to resonate with other companies, whether they are currently in dialogue with the FDA or anxiously awaiting their own letter from the agency. (A good bet for a future FDA letter: companies offering direct-to-consumer telomere measurement tests, several of which drew criticism in recent weeks for over-inflating the ability of telomere length to predict individual longevity.)
The Latest in DTC Offerings. American International Biotechnology Services (AIBioTech), another company to receive a recent DTC letter from the FDA, found itself under the FDA’s microscope thanks to its Sports X Factor Genetic Athletic Assessment Test. The test purports to “reveal a person’s genetic athletic performance indicators as well as the potential for several risk factors,” including the risk of “negative results after an injury to an individual, such as a concussion,” or the possibility of an “undiagnosed heart condition.”
AIBioTech is not the first company to attempt to combine genetic testing and athletics – we dissected offerings from Athleticode and Atlas Sports Genetics back in 2009 – but the company’s test has received additional scrutiny due to its inclusion of markers for the apolipoprotein E (APOE) gene. The APOE gene, which comes in at least three different versions or alleles (ε2, ε3 and ε4), has been shown to affect an individual’s risk of coronary heart disease, as well as for developing late-onset Alzheimer’s disease.
Providing APOE results directly to consumers raises a host of delicate ethical and legal issues, as we discussed last month when DTC veteran 23andMe unveiled APOE reporting as part of its standard service. AIBioTech has drawn criticism for largely ignoring these issues in offering its athletics-focused test. (This in addition to criticism of the scientific validity of the claims offered up by AIBioTech and other athletic genetic testing companies.)
The inclusion of APOE in the AIBioTech test also highlights one of the frequently discussed challenges of trying to regulate any genetic test – whether DTC or otherwise – on the basis of its claims or intended use. As the proliferation of genomic data continues and our collective genomic understanding grows, the number of genetic markers which exert influence upon multiple traits of varying significance is likely to rise.
APOE’s implication in both heart disease and Alzheimer’s is a classic illustration of the phenomenon of pleiotropy – the ability of a single gene to influence multiple phenotypic traits – although it is hardly the only pleiotropic gene (genes responsible for sickle-cell disease, PKU and albinism, among others, also exert pleiotropic effects). As additional pleiotropic biomarkers are identified, the notion that genetic tests should be evaluated based on their intended use is likely to come under increasing pressure. How, for instance, will regulators address a genetic ancestry test when one or more of the biomarkers employed by the test could be used to predict an individual’s genetic risk for a serious disease? While this dilemma is not a conceptually new one, as the regulatory framework for genetic testing evolves, the nonlinearity of gene-trait relationships seems likely to pose a significant challenge for regulators, clinicians and companies who would prefer the ability to provide precisely targeted genetic test results to individuals.
Regardless, the issue seems likely to be mooted in large part at the point in the not-too-distant future where individuals have routine and early access to complete whole-genome sequences. Then the issue will no longer be whether or how to provide data with the potential to help predict multiple phenotypic effects, but how to exert any control whatsoever on the interpretative tools available (including, perhaps, IBM’s Dr. Watson, which is now reportedly “as good as the smartest second year med student”) to individuals who have access to complete and portable personal genomic data.
Are You Ready for RUO? Finally, one very recent development of potential significance to the DTC industry is the applicability of the just-released FDA draft guidance for research-use-only (RUO) and investigational-use-only (IUO) in vitro diagnostic products.
Published earlier this week, the RUO/IUO guidance clarifies the rules for marketing and commercializing diagnostic products under the widely used RUO and IUO exemptions, which allow device manufacturers to avoid submitting their products under the FDA’s rigorous medical device approval pathway. (For more see this article in Pharmacogenomics Reporter.)
The draft guidance, which is open for public comment for the next 90 days, is a significant event for manufacturers of laboratory developed tests (LDTs), many of which are thought to incorporate RUO/IUO components despite being offered for clinical or diagnostic purposes. Expect LDT manufacturers and industry groups, such as the Association of Molecular Pathology and American Clinical Laboratory Association, to have plenty to say on the RUO/IUO draft guidance before the comment period expires at the end of the summer.
Less clear is what impact the RUO/IUO guidance might have on the DTC industry. As with LDTs, it is not known exactly how many DTC genetic test providers utilize RUO/IUO components in their products. However, the pairing of an RUO or IUO device with a DTC product certainly occurs. For instance, 23andMe utilizes the Illumina OmniExpress Plus in its popular DTC service. The OmniExpress is part of a family of popular DNA microarray products offered by Illumina and labeled for research use only (pdf).
This exact issue appeared last summer when the FDA included Illumina in its first batch of DTC letters following the Pathway/Walgreens dust-up. From our coverage last June:
Of all the companies receiving letters, Illumina is the only one not to offer at least one product directly to consumers (the company does offer a commercial whole-genome sequencing service, although that product requires the participation of a healthcare professional). Illumina’s letter notes that the company has “received FDA clearance or approval for several of its devices,” but not for the HumanHap550 array. “Yet Illumina is knowingly providing the HumanHap550 array to 23andMe and deCODE Genetics for clinical diagnostic use without FDA clearance or approval” despite its being labeled “For Research Use Only.” As Gutierrez notes, “…Illumina has to follow the law, and they are aware that the chips are not being used for research only.”
Although the specific product has changed over the past year, the fundamental issue has not. The new RUO/IUO guidance, if approved in its current form, places a heavier burden on providers of RUO/IUO devices to “not sell such products to laboratories they know use the product for clinical diagnostic use.”
Which brings us back to one of the fundamental tensions concerning DTC genetic testing services: whether such services are intended, at least in the eyes of the FDA, for clinical diagnostic use. While the FDA and DTC genetic testing companies may not agree on the reasonably intended uses of these services, the FDA’s draft RUO/IUO guidance threatens to insert another key player – platform technology providers, like Illumina – squarely into the middle of the DTC debate.
If the FDA continues to maintain that the genetic testing services offered by 23andMe, Lumigenix and other DTC providers are intended for use in clinical and/or diagnostic testing, the Illuminas of the world will find themselves with a difficult choice to make. They will likely be forced to (a) defy the FDA’s RUO/IUO guidance, (b) stop selling their RUO/IUO devices to DTC companies or (c) attempt to shepherd their RUO/IUO devices through the FDA’s resource-intensive medical device approval process.
Or as the FDA puts it:
FDA is aware that laboratories sometimes use IVD products labeled RUO in clinical diagnosis and that many manufacturers, importers, and distributors of IVD products labeled RUO are also aware of such use. Manufacturers who label their IVD products: “For Research Use Only. Not for use in diagnostic procedures,” should not sell such products to laboratories that they know use the product for clinical diagnostic use. If a manufacturer learns that a laboratory to which it sells its RUO-labeled IVD product is using it in clinical diagnosis, it should halt such sales or comply with FDA requirements for IVD products, including premarket review requirements, if applicable.
If it were only DTC companies utilizing RUO/IUO devices, manufacturers like Illumina might simply stop selling those devices in light of the small (current) size of the DTC industry. But because those same devices also appear to be used by a number of LDT manufacturers, who represent a much larger market, there is a strong possibility that RUO/IUO device suppliers like Illumina will seek to obtain FDA approval for those devices, hopefully while working with the FDA to keep those devices on the market in the interim to avoid interruptions to patient care (or, in the case of their DTC customers, consumer product supply).
Regardless, the next few months will bear close watching to see whether the FDA’s attempt to more strictly enforce RUO/IUO labeling results in any significant shift in the relationships between DTC genetic testing companies and their technology suppliers.
What’s Next for DTC? Remember that it was roughly this time last year when Pathway Genomics’ failed partnership with Walgreens kicked off a busy summer of DTC (and related) activities at the FDA and on Capitol Hill. Up until this week, Washington had been relatively quiet since the end of last summer, at least in public. While the RUO/IUO situation bears watching, there are some who expect this summer to bring further fireworks on a par with last year.
While several important initiatives continue to loom as possibilities – including the FDA’s long-anticipated LDT guidance and new diagnostic-focused legislation from Congress – with 2011 nearly halfway gone I continue to think that the prospects for industry-wide regulation of DTC genetic testing in 2011 remain dim.