Should Personal Genomics Be Regulated?

I read recently about the cease and desist letters sent to 23andme and other personal genomics companies selling tests directly to consumers. 23andme has responded, saying that they agree with the ultimate need for regulation, but that harnessing the consumer internet for personal genomics is a really valuable scientific tool.

I have to say I find myself doubtful about the urgency of this regulatory move. It smacks more of the hand of the AMA, an entrenched industry trying to make sure that the new tools of genetic testing remain under the thumb of doctors, than of true consumer protection. You have only to walk into Whole Foods to encounter a multi-billion dollar industry of supplements making all kinds of dubious health claims, which is completely unregulated. Why pick on personalized medicine, which has way more substance, and at least so far, way more care in the types of claims it makes?

I’m a happy 23andme customer, and it’s hard for me to see how the information they provide consists in any way of medical advice that should be subject to regulation. They are very clear to mark the scientific status of any genetic studies they report on, and never begin to presume to diagnostics. In many ways, the service is a kind of “RSS for the genome,” feeding you the results of the latest scientific studies that might be relevant to you. You can find these same studies simply by googling for them. 23andme simply says “this might be relevant” using data that you’ve provided about yourself via your genetic sample.

The benefits of what companies like 23andme are doing is enormous. Once you understand even a tiny bit about how genes affect our response to drugs, you realize just how flawed many clinical studies are. If people have different genetically programmed response to drugs, what is the right dosage? Is the drug effective? (What if it’s 100% effective for 3% of the population, due to their genetics, while it’s completely ineffective for the rest. Would we ever know?)

Ideally, every clinical study going forward would have a genetic screening component. And new studies should be done on old treatments to correlate their effectiveness with genetic data. But that’s hugely expensive.

23andme (and presumably other similar services) have come up with a very clever hack that will vastly increase the available store of genetic information that can be used to cross check various medical studies. They recruit a large population of those who are merely curious. But once that genotype data is available, they can begin to do surveys of their user population to gather corresponding phenotype data. (That is, matching genetic data with observable characteristics.)

This will be a boon to science. As Linda Avey, one of the founders, wrote on the 23andme blog:

Our first mission is to enable personal access to genetic information and provide a look, through the prism of an individual’s genome, at the flood of research discoveries being published. Our longer-term goal is to utilize a web-based platform that gives individuals the ability to share details related to their personal traits-including diseases they have and how they respond to therapies-uniformly layered on their genetic profiles to start building the evidence needed to drive targeted diagnoses and treatments.

I hope that regulators will seize the opportunity provided by the consumer internet to open up the frontier of personalized medicine. This is our data, the most personal data we have. I don’t see why it should be forced into the straight-jacket of the 20th century medical industry. We tried that with music, and what did that get us?

I’d say to let this area run for a while, and start thinking about regulation if and when it starts to go awry. Right now, the companies I’m aware of are being very careful not to promise anything that needs to be regulated.

tags: , ,
  • You did remember what happened to Napster right?
    The personal genomics industry is the low hanging fruit with not nearly as much money as the nutritional supplement (Not intended to diagnose or treat) industry…..
    Plus, this technology has been allowed to run amuck for 20+ years now without regulation……Unfortunately for DTC companies, regulations are coming now. My advice, train more doctors to be able to handle the medical parts of the genome. I do love your idea of an rss feed though….It is brilliant. My question: “Who does the unbiased intepretation, translating it into normal human language rather than science-speak”


  • After working in the pharmacogenetics field for the last few years, I am extremely skeptical of the layperson’s ability to make meaningful decisions based upon their genetic profiles.

    The statistics involved in properly interpreting genetic data are complicated, subtle, and tricky. Tens of thousands of samples need to be analyzed, and slight biases in experimental design can easily lead to incorrect results. What’s worse, even when statistically valid results are found, the genetic component usually explains only a small fraction of the trait — odds ratios of less than 1.25 are not uncommon, and results regularly fail to reproduce in follow-up studies.

    The problem is, most people are incapable of grasping the relevance of provisional, statistical information. Say you have a mutation that was carried by 20% of the population and was associated with a 10% increased risk of developing early-onset Alzheimer’s in a study of 3000 Ashkenazi Jews. So what? 90% of the people in the study with your mutation *didn’t* show an increased risk. You have no way of knowing if those results are applicable outside of the ethnicity studied, nor can you really know if you match the genetic and environmental profile of the controls. Even if the result is applicable, well, what are you going to do? Alzheimer’s is untreatable, and few preventative therapies exist.

    But the way these results are presented, people will automatically assume “I’m getting Alzheimer’s.” It’s human nature; it’s the way we perceive risk. We magnify risks outside of our control and minimize risks within our control. We assume genetics is destiny, despite the mountains of evidence showing the influence of environment and lifestyle.

    Sure, 23andme has invented a great “hack” to collect lots of genetic data on the cheap. The potential benefits to research are not to be underestimated. However, I am dubious of the “service” they provide to consumers.

  • bioinformer

    (sigh) That website and the people promoting it are selling you a fallacy. I hold a degree in Molecular Biology and I know that the reams of data being released from studies, in articles and on the web are the normal scientific process of discovering how all these genes and sequences work. The lay person at home has next to no chance of gaining anything useful from home DNA analysis using substandard tools and lacking in statistical analysis techniques. Technicians trained in these techniques with years of knowledge get it wrong all the time because it’s insanely complicated and there are dozens of factors. All this site is going to do is feed the hypochondriacs of the world into believing that they have even more to worry about. Anyone telling you you’ll be able to use these tests successfully might as well be convincing you that you really can operate on yourself and be successful. A novelty tool of minor interest at best.

  • bioinformer, zenkat –

    Doesn’t sound like you’ve actually looked at the site. I don’t see 23andme feeding me anything to worry about. They tell me things like: you share these genes with this brother, those with that sister. It’s a game of exploration. Not much substance yet, but provocative and interesting.

    A lot of people fault the site because it doesn’t make the easy claims that you attack as a straw man — they want simplistic diagnoses, and the site is careful not to give them.

    Why is this different than saying, “My family history is x, so I have an increased risk?” Now you can say, I have this gene, so I have an increased risk, or I don’t, so perhaps I have a decreased risk.

    Would you keep people from knowing their family history as well? I shouldn’t be able to ask my mother if she knows what my grandmother died of because some doctor thinks I might misinterpret the information?

    Priesthoods always like to say that information is too complex for the layman. That’s the best way to keep their priesthood. Meanwhile, that allows true snake oil salesmen to prey on those who are kept ignorant.

    I’m old enough to remember stories of doctors who didn’t tell patients they were dying. I had one friend whose young wife died of cancer with him at her side reassuring her that she would get better (because the doctors told him it was better to keep the information from her.)

    If there are genetic tests that are appropriate to be prescribed by doctors, I’m sure they will be. The availability of this general data doesn’t take away from that.

  • Tim: “You have only to walk into Whole Foods to encounter a multi-billion dollar industry of supplements making all kinds of dubious health claims, which is completely unregulated.”

    Tim, That’s not true. Health foods and supplements cannot make health claims. They use very careful wording to imply health benefits, without actually making the claim. I think what they get away with is shameful, but what they do is regulated and legal.

    As regards the cease and desist. This comes on the heels of the UK crackdown on genetic testing and diet recommendations. I’ve no doubt that this is the primary target of the States. 23AndMe could unwittingly support that industry, by either running paid ads against the interpretation pages or by supplying the data for the health food/supplements industry to make their claims. Clearly regulations to prevent that need to be made.

    Unlike some of the other commetators, I do think that genomic testing is a useful service. It will eventually encompass both genetic testing for diseases that are already validated e.g. Huntingdon’s, and provide the data for more extensive research, far exceeding even DeCode’s efforts. However, I do share the concern that interpretation is still in it’s infancy and that we must not forget that health and disease is both genetically and environmentally determined. Having said that, I think we are at the beginning of a new era where your genetic makeup will become part of the information required for treatment, whether prophylactically based on predisposition or on response to drug therapies (personalized medicine). Also, I could easily see the ‘genomic library’ becoming an indispensable research tool in this area if the data were public.

  • Stephen Murphy –

    Glad you like the line “Rss for the genome.” It’s actually not mine, but Craig Venter’s. I believe he said something like that when I interviewed him on stage at the Web 2.0 Summit last fall. That’s why I put it in quotes. Should have actually worked the attribution into the piece as well, but it seemed to me that the meme was already out there in the world in a pretty general way. That really is how 23andme works — their gene journal says, “here’s some new research you might find interesting,” just like an RSS feed, which is just how Venter described his own experience, albeit with a much deeper data set and more knowledge of how to interpret it.

    Meanwhile, continuing the response about whether genetic information ought to be under lock and key, looking to see if Venter’s formulation was cited anywhere I could link, I found a nice bit in an interview he did last year for MIT’s Tech Review:

    “A key part of the message here is that people should not be afraid of their genetic codes or afraid to have other people see them. That’s in contrast to the notion that this is dangerous information that should be kept under lock and key. We’re not just our genetic code. There is very little from the code that will be 100 percent interpretable or applied.”

    He takes the opposite view from commenters here, but for the very same reason. This isn’t absolute data, and keeping it under lock and key reinforces the bad idea that it is.

  • For me, there are two points to consider.

    I am confused as to why a ‘layman’ cannot understand a 23andme result. This strikes me as illogical. Look at it like the ultrasonic picture that a pregnant womans has taken of her baby. No, the mother won’t understand all the detail, but that is not the point. This is not about the detail, it is about the general impression. It is about curiosity. The same is true for genetic analysis. You are curious. You want to know. It will only be when you have some serious problems that you will be able to recognise them. When you do, you will go and see the doctor.

    This is ‘my’ genes we are taking about. I have a right to know. Why do we need to regulate whether or how I can look at my data? I can actually measure all my phyical properties (size, weight, blood pressure, blood sugar, etc etc) myself. Why should people have a say as to what I can do with my genetic data? Or how I have to obtain it?

    @Tim: I agree, I would give people free access to it. We should deal with problems if/when they arise.

  • N/A

    I cant help but to partly agree with the Napster comparison, but not the way the first poster means it. Napster fundamentally changed the way we share information, and helped shape the way that we use the internet today. They, and other P2P services became over-regulated (remember the RIAA suing students?), and as a result caused a massive backlash from consumers, resulting in a huge drop off in ‘physical’ music sales, paving the way for a company like Apple to sell billions of songs via iTunes. I think you will see similar things happen in this space, the regulators will undoubtedly (if they havent already) cause the demise of some players, others will pave the way for the general public to be able to access the information they have the right to access, without being ‘forced’ to involve a doctor.

    For what it’s worth, I also don’t think that companies like Navigenics or 23andme get credit for the incredible teams they have assembled (just go to their websites and lool). Their combined knowledge of this field far outweighs many of the so called ‘experts’ chiming in on the subject.

  • There’s only one point of argument here, Jens stated it, but I think it needs to be emphasized:

    Do you think I have the right to know my own DNA sequence or not?

    I worked in the industry for 8 years, and of course it’s full of hyperbole and claims of causality based on the slimmest of correlations. I still believe the ethical path is to let people see their DNA and then tell them there’s no useful interpretation.

    Would you deny people the ability to look at their own palms because palm-reading is inaccurate?

    The last question was rhetorical but this is a real question — what information should I not be told about myself?

  • Well, let’s go over it then.

    Something as critical as medical testing deserves a bit of, if anything, over-engineering. Thus, traditionally, testing labs do not retain data. This is the first sin of 23andme for they built a business model around retaining data.

    Second, human subject medical experimentation — especially for something as intimate and implicating as genetics testing (the decision of one impacts the entire family) — well, human subject testing is supposed to require lots of hoops and jumps, not just clicking on a TOS button and giving a credit card. Indeed, when Harvard contemplated collecting similar data from me the first idea for an offer they put forward was “top tier medical insurance for life, gratis” rather than 23andme’s invoice for $1,500 (or whatever the exact amount is). (Having gotten close to some of the researchers and sussed them out a bit, guess what: I closed those negotiations with “no way in hell”.

    The A.G. office is going after 23andme on technicalities but, my interpretation, going after them because there are much larger problems with this “new eugenics” business.


  • I *love* your palm reading analogy, Toby! Totally right on.

  • Seems like there might be a First-Amendment free speech issue here. True, commercial speech doesn’t enjoy the same freedom from regulation as political- or religious speech. But on the facts discussed, it’s hard to see a justification for limiting 23andme’s ability to tell people about their genes, even for profit.

  • I’m not a fan of the crude and heavy-handed rules the California Department of Public Health is trying to enforce, particularly their clumsy attempt to force medical doctors into the genomics process. At the same time, though, I think technophiles and early adopters tend to understate the consequences of an unfettered personal-genomics market.

    As I’ve written at greater length over at BNET, some sort of collision between health regulators and personal-genomics companies was all but inevitable. First, although some genomics outfits now claim that SNP scans and the like are mere “information services” with no medical utility, that’s already a debatable proposition — and it will grow increasingly tenuous as more and better association studies yield data.

    Second, it’s not hard even now to find companies hawking questionable genetic tests or genome scans directly to consumers, sometimes specifically to con people into dubious nutritional-supplement regimens intended to “counteract” supposed genetic predispositions. This sort of scammery is only likely to get worse over time. Finally, there’s the potential impact of mass-market genomic testing, particularly of the unscrupulous variety, on the “worried well” — and thus on overall healthcare costs.

    In other words, now is as good a time as any to fight for the smartest set of rules possible, because the regulators aren’t going away. Ideally, the resulting framework would respect the potential of genomics while also acknowledging that some of the issues here go well beyond any single company — much less the individual “right” to see one’s DNA.

  • Really good points, David. I certainly don’t see any reason for regulators not to crack down on the kind of shady promoters that you describe. But they started with cease-and-desist letters to folks that definitely don’t fall into that category.

    That seems off the mark to me.

    If there are indeed folks hawking particular supplements based on genetic tests, they are over the edge. But I’d guess these folks have been hawking those same supplements in the past, just without the genetic test as marketing hook. So I’d say, crack down on the people making health claims, not on my right to get a look at my own genome!

  • What’s fascinating about this debate is the transparent double standard being applied.

    Let’s see! The complaints are (1) that information tying genes to disease is imperfect and sometimes contradicted, and (2) ordinary people are prone to misunderstand and misuse the information and make cosmically stupid decisions.

    On point 1, the good news is that there is no genetic determinism. Having a gene or not having a gene doesn’t automatically translate into a specific phenotype. (Setting aside those famous autosomal recessives). Environment shapes how that gene (or genes for polygenic traits) translate into phenotype.

    But the mainstream has no hesitation in telling people that if they don’t lose 30 pounds, they will get diabetes or have a heart attack. Or that women should eat lots of colored berries, green tea and flax to prevent breast cancer. Does anyone seriously believe that if every women ate a breast cancer diet, that none would get breast cancer? Or that if no-one ate that diet, they would all get breast cancer? Of course not. It’s just a statistical shift. And what about second hand smoke? Sure the epidemiologic studies show an increased risk of acquiring some lung diseases as a consequence of second hand smoke. In the meantime, you fly to Europe and encounter a cloud of second hand smoke over the continent, but the Europeans tend to outlive us.

    On the misuse of information, while it is true that there are some cosmically stupid people, most people who are interested enough to learn about their health and genetics are able to absorb collateral information to help them understand things. In terms of physicians, very few have any real training in genetics. And most of that training is superficial, and often outdated. While it may be desirable to have physicians learn more genetics, say in medical school, which parts of the 4 year curriculum should be eliminated to include genetics? The problem is that knowledge in every medical field is exploding, which makes packaging it into a medical curriculum even more difficult.

    Finally, most of our genome works around normal functions, not disease. Genes influence our personality, physical abilities, how we process nutrients, energy management and a host of other everyday life traits. Do we want to require a doctor’s order to find out if you have a gene that relates to athletic ability or how food tastes? If so, then maybe we want a doctor to prescribe which movies we should see, or whether it is ok to go to McDonald’s or have a cup of coffee.

    Craig Ventner said in a speech recently that consumer genomics “democratizes” genomics. What’s wrong with that?

    Fredric Abramson

    So that you know my position, AlphaGenics is dedicated to packaging information about normal genetic functions in consumer friendly ways, so that people can make better, more informed lifestyle decisions right now.

    We recognize that the knowledge base is the tip of the iceberg. We believe that making genetics a universal part of everyone’s everyday life will drive the scientific research to uncover the remainder of that iceberg more quickly and more accurately.

  • James

    I doubt the AMA has anything to do with the warning letters. Regulation of “genetic testing,” in particular DTC testing is a hot topic today. There are a number of public policy types and genetic counselors who are agitating for this. That being said, claims that 23 and me will perform valuable clinical research seem foolish. They are simply selling a medically valueless product over the internet. The Company is not prospectively selecting patients, systematically gathering data, or controlling for variables. I don’t have strong feelings about these types of Companies – if people want to waste their money, why should I stop them? On the other hand, I feel quite certain the market will weed them out.

  • The disintermediation of music by P2P folks like Napster might be instructive for understanding what is happening to the consumer genomics industry. Especially in the sense that we should not be surprised by knee-jerk and over-bearing reactions by those with calcified business models and interests.

    As I have written elsewhere, there are a few analogies from the history of medicine itself which may provide a better roadmap for how the shake-up will play-out. I don’t think the fate of Napster is very helpful in this regard.

    Insulin injections and pregnancy tests are both enduring examples of disintermediation in medicine. In both cases, physicians had to take a back seat and cede control to patient self-management, but for different reasons.

    For pregnancy testing, the *possibility* of disintermediation was due to technological advancements, i.e. urine dipsticks rather than rabbits. However, the realization of self-management for pregnancy testing did not come without a fight. Physicians were at first reluctant and discouraging, and even publicly shamed women who would consider testing at home rather than the clinic. “What were they trying to hide?”, they asked.

    For insulin injections, the reason was economic. If insulin injections were needed by diabetics only once a year, rather than 2+ times a day, physicians would still control the syringe.

  • ds

    Of course, I have the right to pay a company to tell me what bases I have at which positions!

    And the company of course also has a right to post links to scientific publications in which disease/eye color etc. associations of these positions are reported!

    I don’t see any need for a physician or the government in this process!

    A happy deCODEme customer :)

  • @Tim: Although I don’t think the problems of laissez-faire personal genomics end with shady promoters, I’m happy to take agreement where I can find it.

    I did want to add, though, that one of your original arguments — that 23andMe and their compatriots are “vastly increas[ing] the available store of genetic information that can be used to cross check various medical studies” — strikes me as an overstatement. Yes, it’s clear that 23andMe would like to sell access to its database, as that’s probably the only way it’s ever going to make money. But the usefulness of that data is likely to be pretty limited, largely because using it for association studies will require investigators to rely on customers’ self-reported phenotypes.

    The story might be different if the company could convince its customers to turn over their complete Google Health records, but I suspect that isn’t going to happen any time soon. And even there the self-reporting issue would be a problem absent medical review by investigators — who would most likely be M.D.’s, and then there you are with those pesky doctors once again :-).

    @Jason: Interesting analogy. But diabetics still need a prescription for their insulin, right? In other words, the disintermediation only went so far in that case, and I suspect that’s likely to be true here as well.

    I’m not at all opposed to disintermediation in general, of course. I’m all for new technology and individual initiative cracking open hidebound bureaucracies and cabals. I just don’t see it going as far as genomic absolutists seem to, and as the medical utility of these services increase, I expect doctors and genetic counselors to get with the program as well.

  • Just a cross-reference link to a related article at the Tech Review:

    From the intro:

    “It’s now possible to go online and mail-order genetic tests that will predict your genetic risk for common diseases. But because the meaning of the genetic risk factors flagged–how they boost risk for disease–is not yet clear, such tests have received major criticism. In many cases, it’s unclear how to use the findings to improve an individual’s health … A new $31 million project aims to address that issue by revisiting large epidemiological studies encompassing tens of thousands of people whose medical status has already been well documented.”