DTC Genetic Testing and the FDA: is there an end in sight to the regulatory uncertainty?

Editor’s Note: This was first published at Genomes Unzipped and was co-authored by Daniel MacArthur and Luke Jostins. Genomes Unzipped received 12 free kits from Lumigenix for review purposes, and Dan Vorhaus has provided legal advice to the company. Genomes Unzipped plans to release a full review of the Lumigenix service in early July.

Last month three direct-to-consumer (DTC) genetic testing companies opened their mailboxes to find a slightly ominous but entirely expected letter from the FDA. The three recipients (LumigenixAmerican International Biotechnology Services and Precision Quality DNA) received substantively equivalent letters, with the FDA warning each company that its genetic testing service “appears to meet the definition of a device as that term is defined in section 201(h) of the Federal Food Drug and Cosmetic Act,” and that the agency would like to meet with company representatives “to discuss whether the service [they] are promoting requires review by FDA and what information [they] would need to submit in order for [their] product to be legally marketed.”

Translated from bureaucratese, that means that the FDA views these services as ones that may need to be formally reviewed by the agency and either approved or cleared before they can be legally sold. The FDA letter asks each company to describe its service and to explain either (1) why it does not require FDA approval or (2) how the company plans to pursue such approval.

This is a strategy that the FDA has pursued with a growing cadre of DTC service providers. These letters (currently 23 and counting1) represent the only public and company-specific actions the agency has taken to date with respect to DTC genetic testing. While many DTC letter recipients are engaged in dialogue with the FDA, those conversations have occurred beyond the public’s view. Until now.


Ending the regulation guessing game?
For more than a year, the FDA has dealt with DTC genetic testing providers by mailing (and publishing) an initial letter followed by the initiation of a private dialogue and a company-specific regulatory determination. The agency has yet to publish, or even to propose, anything resembling industry-wide regulatory guidance for current or prospective personal genomics companies.

For their part, personal genomics companies have been reluctant to publicly disclose the nature of their conversations with the FDA. Recently, however, a new approach appears to be emerging, at least on the company side.

Last month, a recent DTC letter recipient (Precision Quality DNA) published a strongly worded response to the FDA on its website. Another letter recipient, Lumigenix, soon followed suit, launching a new corporate blog with the publication of its own more measured response to the agency. Each company undoubtedly has its own reasons for bucking the prevailing trend and choosing to take its conversation with the FDA into the public square. There should be little doubt, however, that the FDA’s current company-by-company approach to personal genomics regulation, which has left the industry with a considerable measure of uncertainty, was a major contributing factor in each decision. As Dan and his colleague Allain Andry noted last summer, the effects of that regulatory uncertainty can be substantial.

They include:

Current and prospective DTC genetic testing providers alike are no doubt burdened by some or all of these challenges. Lumigenix, in particular, appears hopeful that a more public DTC discussion might help to lessen the effects of regulatory uncertainty, noting at several points in its response to the FDA the company’s desire to work with the agency to develop “a clear and reasonable system of oversight for the emerging field of personal genomics.”

The challenge of “clinical” DTC claims. Perhaps the greatest area of current DTC regulatory uncertainty concerns the ability of companies to provide interpretations of personal genomic data with potential clinical or medical significance. The FDA has consistently maintained that its regulatory interest lies first and foremost with clinical genetic tests, with a test’s intended use determining whether it qualifies as clinical.

The FDA’s emphasis on clinical genetic testing has presented personal genomics companies with a dilemma: offer medically relevant personal genomic results in response to consumer demand and thereby risk stricter scrutiny from regulators, or attempt to cater to regulators (but risk losing customers) by removing or deemphasizing results that could be construed as clinical.

Personal genomics companies have deployed a variety of solutions in response to this dilemma. 23andMe, for instance, has simply braved regulatory ire by continuing to offer tests for the BRCA breast cancer risk variants, pharmacogenomic response and other serious disease mutations. Pathway Genomics responded to its own FDA letter by promptly eliminating the ability of consumers to purchase its product without physician involvement. Carrier testing company Counsyl avoided a letter entirely by dropping DTC marketing as soon as the FDA began to make serious regulatory overtures.

Lumigenix, for its part, has thus far taken an intermediate approach, steering clear of reporting on variants with unambiguous clinical relevance while maintaining DTC access to its service. In its response to the FDA, Lumigenix emphasizes that the company “strongly believes that individuals should have the right to access their own genetic information,” but explains that Lumigenix has opted to exclude certain information from its current service to avoid any clinical confusion:

In order to ensure there is no doubt about the educational purpose of our service, Lumigenix’s current service intentionally excludes certain categories of genetic tests. For that reason, Lumigenix’s does not currently include in its customers’ genomic reports results from genotype data known to be associated with or to indicate:

carrier status for a recessive disease (e.g., Cystic Fibrosis or Tay-Sachs disease);
pharmacogenomic status related to an individual’s likely response to certain medications (e.g., Warfarin or Clopidogrel); or
a serious or untreatable illnesses with a large genetic component (e.g., breast cancer, Huntington’s disease or Alzheimer’s disease).

We understand that some individuals desire such results for their informational value. But we also acknowledge that the risks associated with personal genomics, including the risk that an individual will make an important medical or other decision without first consulting a healthcare professional, are not equal across all categories of genetic tests.

For that reason, Lumigenix has decided to focus its current service on providing personalized genetic information pertaining to genetic ancestry, non-medical traits and conditions, and certain relatively common medical- or health-related conditions that tend to be influenced by many genes and include a substantial environmental component.

It’s worth noting that Lumigenix is still early in its conversation with the FDA and has not yet had the same length of time to respond (e.g., in the form of modifications to its service or business model) as earlier DTC letter recipients, including 23andMe and Pathway Genomics. Lumigenix’s response does, however, hint that changes may be in store: the company applies the adjective “current” in describing its service eight separate times in its response to the FDA.

The range of approaches taken by personal genomics companies on this issue alone reflects a much broader uncertainty about the industry’s regulatory future in the hands of the FDA. The agency’s reluctance to publicly pursue a comprehensive personal genomics regulatory framework is understandable, particularly in light of the rapid pace of scientific and technological innovation and the paucity of data about DTC genetic tests and their affect on consumer behavior. Still, for so long as the FDA continues with its current private, company-by-company regulatory approach, personal genomics innovation and investment are likely to remain hampered by uncertainty.

A glimpse into the regulatory future. Interestingly, despite Lumigenix’s focus on non-medical data as part of the formal interpretations it provides to its customers, the raw data generated by the company’s genome-wide test contains plenty of medically relevant genetic variants. For example, there are at least 196 sites on the chip that match the position and sequence of known Mendelian disease mutations, including 12 in the BRCA1/2 genes and 6 in the cystic fibrosis gene CFTR2. Lumigenix simply does not report on or interpret these variants, although customers can choose to explore those variants on their own, including through the use of free software such as SNPedia’s Promethease.

This approach to medically relevant personal genomic data, at first blush needlessly confusing and inefficient, is unsurprising in light of the FDA’s insistence that their regulatory target is not genomic data but the claims – particularly clinical or medical claims – made on the basis of those data. Last August, for example, reporter Mary Carmichael and Dr. Elizabeth Mansfield, Director of Personalized Medicine for the Office of In Vitro Diagnostic Device Evaluation and Safety (OIVD), had the following exchange:

[MC]: I want to move on to whether the issue with direct-to-consumer is actually providing data to people, or is it the interpretation algorithms these companies are using? So, would a company need to be approved just to provide a raw SNP list to people?
EM: They would if they made medical claims about that data. If they don’t make any medical claims about that data, then they’re free to provide information as far as we’re concerned.

Over the past year, nothing has happened that would suggest that the FDA has revised its policy on this point. This bodes well for both DTC companies and their customers because, barring a striking reversal by the FDA, high-quality personal genomic data appears likely to remain readily available, including via direct-to-consumer channels. Whatever else the FDA may have in store for the personal genomics industry, the availability of raw genomic data should enable DTC companies to remain on the market in some form.

Still, the increasing availability of raw personal genomic data will itself soon pose a challenge for the FDA. Already widely accessible and inexpensive, personal genomic data will soon transition from SNP chips to whole-genome sequences. As data proliferates alongside increasingly numerous and sophisticated publicly available software tools (like PrometheaseSNPTips and Interpretome, the topic of a recent post) used to mine those data, the FDA will find that focusing on all-inclusive providers of DTC personal genomics services is insufficient. As the separation of testing (genomic data generation) from interpretation (genomic data analysis) accelerates, the FDA will be faced with a growing array of personal genomics service providers, many of whom are likely to provide software-only tools and will see no pressing need to operate from within the United States (and within easy reach of the FDA).

How the agency responds to the inevitable expansion and diversification of the personal genomics industry – e.g., with an expanded letter-writing campaign or a concerted effort to develop flexible and forward-looking industry guidance – remains to be seen.

_____________________________________________

1. The count: Pathway Genomics (May 2010); 23andMe, Navigenics, Knome, deCODE Genetics, Illumina (June 2010); Graceful Earth, SeqWright DNA Technology Services, Interleukin Genetics, DNATraits, CyGene Direct, Consumer Genetics, Matrix Genomics, The Genetic Testing Laboratories, Sequenom, EnteroLab Reference Laboratory, BioMarker Pharmaceuticals, DNA Dimensions, HealthCheckUSA, easy DNA (July 2010); Lumigenix, Precision Quality DNA, American International Biotechnology Services (May 2011).

2. Analysis by DM: I looked at the overlap between the SNPs included in my raw data from Lumigenix (available for download here) and the disease mutations in the full version of the Human Gene Mutation Database (available only via an academic collaboration or a license fee, unfortunately). I counted positions where both the position and both alleles matched. Note that this approach won’t detect disease-causing insertions and deletions, only SNPs.

Filed under: Direct-to-Consumer Services, General Interest, Genetic Testing/Screening, Genomic Policymaking, Genomic Sequencing, Genomics & Society, Industry News, Legal & Regulatory, Pending Regulation
Tags: , , , , , , , , , , , , , , , , , , , ,