The U.S. Food and Drug Administration issued a warning letter to Ann Wojcicki on Nov. 22 that will prevent her company from marketing its sole product for the foreseeable future. No, the company doesn't manufacture pharmaceuticals, nor will you find an associated ticker. It is, however, leading the charge to make your personal genetic information -- the most important information you own -- more accessible to you. The company is 23andMe and the scrutinized product is the Saliva Collection Kit and Personal Genome Service, or PGS.
23andMe hasn't been fully cooperative with the FDA over the years, but consumers have certainly lost -- at least for the moment -- an important tool for accessing their personal genetic information. How the situation unfolds could have profound effects on the personal genome kit market, which has been largely unregulated to date. Did you just lose the right to access your personal genetic information for good?
What's the fuss about?
First, we need to explain 23andMe's business. The company offers a genetic analysis kit for $99. Once it's mailed to your home, you spend a few minutes filling a tube with DNA-rich saliva before carefully repackaging the assembly and mailing the kit back to the company for analysis. After the laboratory completes its analysis of your DNA sample, you can login to your personal 23andMe account to see hundreds of interesting genetic correlations in health and ancestry between your sample and known-genetic correlations from known scientific literature.
Results are based on SNP genotyping, which isn't as comprehensive as full genome sequencing. Nonetheless, the information is what makes you, well, you. So it only seems fair that you should have access to it. However, unless you're a trained geneticist that can successfully and swiftly derive meaning from a pile of genetic information in a specially designed database, then all of those funny letter combinations that make up your genes are pretty worthless. That's where companies such as 23andMe come into play.
The company takes your genetic information and compares it against the vast amount of scientific data available. It isn't really doing anything novel except interpreting genetic data from a software program and reporting it back to you in a neat and organized package. That's still pretty valuable and interesting. And I ought to know, as I paid for the test this summer.
Your ancestry (what parts of the world your genome came from) and traits (eye color, lactose metabolism, and the like) can be easily deduced from years of established scientific research. It seems obvious that genes encode for eye and hair color, but they also play a role in determining your susceptibility to diseases and how you metabolize drugs. 23andMe reports these results in the following categories:
My health risks are reported as estimated percentages when scientific studies of sufficient size have been conducted. Otherwise, my risk is classified as elevated, decreased, or typical.
For instance, I have a 24.9% risk of developing prostate cancer in my lifetime, compared with just 17.8% for the average male, according to the company's analysis of 12 reported genetic markers for the disease. Navigating to the page for prostate cancer provides further facts and statistics on the disease. "The heritability of the cancer is estimated to be 42%-57%," reads the opening of one section, explaining that environment also plays an important role. That's an important thing to consider when digesting your results, as is the fact that 23andMe's test is not designed to diagnose disease.
The company does advertise that results could be used as a "first step in prevention," which could spur consumers to talk to their doctors about investigating their health risks in more detail. That could actually be a good thing for the nation's health. For instance, my elevated genetic risk for prostate cancer means I should take screening seriously when I get older, although I shouldn't be too concerned about it until I speak with a trained medical professional. The FDA doesn't see it in the same light, however. From the letter:
FDA is concerned about the public health consequences of inaccurate results from the PGS device; the main purpose of compliance with FDA's regulatory requirements is to ensure that the tests work.
The truth is, the FDA hasn't proved the company's tests and analyses accurate. You don't have to be too creative to see how arming millions of people with genetic risk factors from unregulated personal genetic tests could cause problems. For instance, my results show that I have increased sensitivity to warfarin, a blood thinner that is prescribed to more than 2 million people in the United States each year. It doesn't really affect me, but if someone who needed to take warfarin for medical reasons had similar results and decided not to take it based on those results, the stakes get raised considerably.
It may seem that the FDA is picking on a small company just for fun, but consumer safety is nothing to joke about. The agency also has the law on its side. According to the Federal Food, Drug, and Cosmetic Act, which was amended in 1976 to regulate medical devices, the PGS is classified as a medical device because it "is intended for use in the diagnosis of disease or other conditions or in the cure, mitigation, treatment, or prevention of disease." Genetic testing wasn't available back then, but neither were laser surgical tools or pacemakers with microchips, which are mandated by the same law.
From that perspective, 23andMe needs to clear its product through Premarket Notification, or the 510(k), process. Such products aren't technically approved by the FDA, but they meet sufficient safety and manufacturing guidelines to be marketed. Unfortunately, the company hasn't done itself any favors in its fumbled attempts to bring the PGS into compliance with the FDA over the years. Two 510(k)s submitted by 23andMe in 2012 are now considered withdrawn in the eyes of the agency. From the letter:
As part of our interactions with you, including more than 14 face-to-face and teleconference meetings, hundreds of email exchanges, and dozens of written communications, we provided you with specific feedback on study protocols and clinical and analytical validation requirements, discussed potential classifications and regulatory pathways (including reasonable submission timelines), provided statistical advice, and discussed potential risk mitigation strategies.
Knowing that part of the relationship makes it awfully difficult to side with the company on the issue or have much faith in management. How the situation is handled now will affect not only 23andMe, but consumers and other companies seeking to empower individuals with their genetic information.
Why does this matter?
23andMe isn't the only company offering cheap genetic tests to the masses. The American Gut Project offers a $99 test that will identify the makeup of your microbiome -- the collection of helpful microorganisms that reside in your intestines -- which has become increasingly important in understanding various diseases. Meanwhile, the crowdfunded start-up uBiome offers a range of tests for sequencing the microbiome starting at $89. Similar to the test offered by the larger 23andMe, both tests are intended for research and educational purposes only. If recent events are any guide, then they may also come under the scrutiny of the FDA.
Are individuals about to lose the right to access their personal genetic information? Regardless of outcome the answer is "no," but the problem is price.
You really can't find a similar product on the market for a similar price. You can pony up quite a bit of money for much more thorough and accurate results, but traditional diagnostic tests are only used for specific medical reasons. They require substantially more money to develop and meet compliance, which means they often come with lofty price tags -- even if they assess only a handful of your 24,000 genes. However, for diagnostic tests to be successful they generally must be paid for through health insurance, which makes their costs largely invisible to consumers.
Myriad Genetics came under intense scrutiny for the costs of its BRCA test -- for determining genetic risk factors for breast cancer -- which cost nearly $3,000 for analyzing just two genes. Genomic Health offers genomic diagnostic tests for breast, colon, and prostate cancer. These and other genetic tests are incredibly useful for determining the likelihood of recurrence and guiding physicians to the best treatment options. Of course, that's what they're intended to do. Thus, they are not a suitable alternative for accessing wider swaths of genetic information.
The PGS 23andMe offers is not intended to diagnose diseases, and the FDA has not proved it accurate. It doesn't offer anywhere near the precision of a diagnostic test for any indication, but then again, that isn't its purpose. So what if the issue at hand had more to do with how the FDA believes the PGS should be classified in the first place?
What can be done to fix this?
While the FDA has been attempting to bring 23andMe's PGS into compliance for years, the decision to put its regulatory foot down now comes at an interesting time in the trajectory of DNA sequencing costs. The primary objective of the Human Genome Project, which sought to sequence the entire human genome, was to spur advances in our understanding of genomics and advance DNA sequencing. We aren't there yet, but we're well on our way. Consider that the cost to sequence DNA (the blue line) has fallen from $1 per unit of DNA in 1995 to an estimated $0.0000001 per unit of DNA today, according to noted venture capitalist Steve Jurvetson:
The advancement has resulted in ultra-cheap technology; technology such as personal genome kits that cost $99. They'll only continue to get cheaper and more accurate, which means they'll become more widespread and more accessible to the general public, which means it's important for the FDA to clamp down on mail-order analysis tests now rather than later.
Then again, it isn't too unreasonable to support 23andMe in its assertion that personal genome kits shouldn't be classified as diagnostic tests. Consider the scenario where the company meets all guidelines the FDA sets and is once again allowed to market its product. Will that behind-the-scenes marketing approval really stop a handful of consumers from abusing the information presented to them? Most certainly not.
Can a compromise be reached between the FDA and 23andMe assuming the company can water down its marketing campaign and clarify that its incomplete tests should not be used as diagnostic tests? Surely, few could have foreseen a future where DNA kits were being mailed around the country by everyday people when the law was amended in 1976.
While I would cheer the decision to make a new, less restrictive classification for current personal genome kits, that doesn't seem likely to occur. The company's test -- and others like it -- is still likely to be classified as a class 2 medical device under the law, which breaks down into three categories. Simple things such as bedpans and dental floss are considered class 1 medical devices, diagnostic tests and hearing aids are class 2, while complex devices that are permanently implanted are class 3. Logically, each class must meet a set of escalating conditions for marketing approval.
That shouldn't be cost-prohibitive for 23andMe, so there's no excuse to not be in compliance. The same may not hold for smaller companies such as uBiome (unless it can crowdfund itself out of that problem, too), which may dampen the progress of kits in the short term. That will change over time as sequencing costs continue to decline. Simply put, the only solution at hand is for any company offering personal genome kits to bring their product into compliance as a medical device.
What does it mean for your personal genetic information?
The issue comes down to two things: intent of the tests and how to classify them. The federal government, top academic institutions, and the health-care industry have invested heavily in technologies allowing for easier genome sequencing in the past decade. The inevitable result of that research will be the breaking down of cost-prohibitive barriers to an individual accessing his or her genetic information. We are still some time away from fully understanding the human genome with great detail and accuracy or cheaply sequencing it in full, but companies such as 23andMe show that day is fast approaching.
It seems silly to classify personal genome kits and the data mining software utilized as a true diagnostic test, so there may be room to make a new medical device category or exemption. Whatever route is taken, know that the FDA is simply taking a proactive approach in dealing with the consequences of this technological advancement. It is by no means the end of 23andMe or your ability to cheaply access your genetic information. In fact, you can look forward to even cheaper tests in the future that will tell you even more about yourself in more accurate detail. How's that for taking the long-term approach?
Editor's Note: A previous version of this article stated that Genomic Health offers genetic tests rather than genomic tests. The Fool regrets the error.