(A reminder: posts on Genomes Unzipped represent the views of individual authors, not the group as a whole.)
Today’s US Congress Committee on Energy and Commerce hearing into the direct-to-consumer genetic testing industry was a vicious affair. Representatives from testing companies 23andMe, Navigenics and Pathway faced a barrage of questions about the accuracy and utility of their tests, made all the worse by the fact that many of the Committee’s members seemed unable to distinguish between the more responsible companies in the field and the scammers and bottom-feeders. (You can read the written testimony of the speakers here, and Dan Vorhaus has a fantastic summary here.)
And the news for direct-to-consumer companies just keeps getting grimmer: the star attraction of the hearing today was a new report from a sting operation by the US Government Accountability Office (PDF), which details the results of anonymous purchases of kits from four DTC testing companies as well as assessments of marketing from 11 other companies approached by the GAO “both by phone and in person” without purchasing kits. While the companies are listed as anonymous numbers in the report, they were revealed in the hearing as (1) 23andMe, (2) deCODEme, (3) Pathway Genomics and (4) Navigenics (the remaining 11 remain nameless).
The report details a litany of complaints – ranging from the flimsy to the serious – about the marketing, reporting and scientific basis for the companies’ operations. Following hot on the heels of the two-day FDA meeting on lab-developed tests (see summaries from Dan Vorhaus here and here) and the recent warning letters sent to 14 more genetic test providers, this is a stunning blow to the nascent personal genomics industry.
Wait, there’s more
And it gets much, much worse: the report includes covertly taped conversations between GAO employees and several DTC companies, which I’ve embedded below. In this video, the company offering breast cancer advice is apparently Navigenics, and the company enthusiastically promoting non-consensual DNA testing for a customer’s fiance is Pathway Genomics. It seems safe to assume that the remaining companies on the tape aren’t members of the first four (reputable) testing companies, but rather of the still-anonymous 11.
Here’s the tape. Brace yourself:
There’s no denying it: that tape is pure gold for the critics of the DTC testing industry. In the first and third clips, a couple of poorly-trained call centre operators at otherwise reputable companies nonchalantly produce the stake that will now be driven into the heart of the DTC industry, over and over again. The remaining three clips appear to depict scam operations – nothing like the products offered by 23andMe, deCODEme or Navigenics – that will nonetheless be conflated in the public mind with DTC genetic testing. Mission accomplished, GAO.
The report itself includes some nonsensical complaints: for instance, the fact that “one donor who had a pacemaker implanted 13 years ago to treat an irregular heartbeat was told that he was at decreased risk for developing such a condition” (the tests provide a probabilistic risk prediction, not a diagnosis). However, others are more serious. For instance, the differences in the risk predictions offered by different companies to the same individual is a problem that has been raised periodically since the industry’s inception; there’s a clear need for industry-wide standards for marker inclusion and background risk figures.
The report also notes that “customers” pretending to be of non-European ancestry were often not informed in advance that their risk predictions were limited (due to the extremely small number of studies on complex disease risk in non-European samples). It’s not the fault of testing companies that genetic research outside of white US and European populations has moved slowly, but warning non-European customers of the limitations of their results should be standard for testing companies.
But overall, the document is obscenely one-sided. It conflates responsible companies offering scientifically valid products with small-time con artists. It ignores the remarkable effort that has been expended on creating intuitive interfaces that allow consumers to grasp complex risk predictions far more easily than anything you’ve seen in a GP’s practice. It ignores the remarkable technical accuracy of the companies’ products, which measure hundreds of thousands of genetic markers with an accuracy over 99.99%. It ignores the fact that the vast majority of personal genomics customers are satisfied with the experience, to the point that reporters seeking to present negative experiences need to exaggerate to do so.
Nonetheless, I’ve no doubt that the report will hit its mark. By this time tomorrow its central claims will have been echoed verbatim by a thousand credulous news agencies around the world, public confidence in the industry will fall, and the regulators will rejoice.
What next?
I only have time for some quick thoughts tonight, so I’m hoping that fellow Unzipper Dan Vorhaus will chime in with his thoughts at Genomics Law Report (update: here they are).
Firstly, there’s no question that the sheer scale and ferocity of this combined inquisition from the FDA and Congress will forever change the face of the personal genomics landscape. Already several previously DTC genomics companies have fled the direct-to-consumer market; Navigenics only offers tests via a “physician or corporate wellness program”, and Pathway dropped its consumer offering at the time of the Walgreen’s incident that sparked this whole mess. It is hard to believe that 23andMe will do the same, although I’ve heard suggestions that the company could drop its disease risk predictions entirely, and instead focus on recruiting participants for its research-based business model via ancestry testing.
Secondly, the momentum seems to be well and truly in favour of the bureaucrats now. The prospect of increased regulation (specifically from the FDA) seemed to be enthusiastically received by the Committee today; there was explicit mention of increased money for the FDA to support such a move. The shape of this regulation is as yet unclear, but I’m now extremely pessimistic about the industry’s prospects of escaping excessive, innovation-crushing regulation in the US.
Thirdly, this is far, far bigger than personal genomics. The FDA is clearly angling to exert its control over a much broader range of tests, meaning its regulatory claws will likely sink into thousands of small non-profit labs performing custom diagnostics for rare diseases, as well as startups seeking to build new technologies for genetic testing. The suffocating effects of this power-grab will be felt well beyond our little circle of personal genomics afficianados.
Finally, I do expect the industry to survive; but if the regulators have their way, it won’t be in the US. Instead, the startups and entrepreneurs that are building the technologies required for the new era of genomic medicine will quietly pack up their tools and move to Singapore and Hong Kong. That’s great, but things are different in Singapore. People will still be buying genetic testing products; but an opportunity to build a scientifically-based, socially responsible personal genomics industry in the US will be gone.
A grim day
On a personal note, today has been a difficult one for me. For all the faults of individual companies, the personal genomics industry has always excited me. Here, for the first time in history, has been an opportunity for individuals to peer inside their own genomes – to trace patterns of inheritance within their families at the raw genetic level, and to engage with the complex, messy reality of modern genetics. Here has been a unique chance for the creation of a community of genetically literate individuals, eagerly watching each advance in genomic research as it revealed new insights into their own past and future. And here, for all its imperfection, was a hint of the shape of the shiny genomic era to come.
I don’t know how much of this will survive the next few months; but if the regulators are allowed to have their way, it will be dismantled piece by painful piece – and that makes me angrier than I’ve been in a long time.
RSS
Email
Twitter
Authors Elsewhere
Hey Daniel–call me an eternal optimist, but once the fury and flurry of the GAO report subsides and Congress moves on to its next topic, the real work will hopefully begin to address some of the valid issues raised today. First, there’s no question that the creation of universal industry standards around genetic risk prediction, if done thoughtfully and carefully, will help build confidence around the information provided by personal genetics companies who adopt such standards. Second, it’s high time for the FDA to create a new, more streamlined–and less costly–regulatory path (as evidenced in this week’s meetings) not only for LDTs but also for comprehensive genome scans that include both analytically and clinically validated markers having unquestionable utility. If FDA officials are willing to keep open minds to entrepreneurial ideas around this subject, all this hoopla could result in actual progress.
Meanwhile, the FTC should get busy looking into the more fraudulent companies. As one official from that body said at a SACGHS hearing a few years ago, they have all the laws they need to prosecute false claims. Have at it!
Hi Linda,
Your optimism is cheering! I guess this is the key question:
If FDA officials are willing to keep open minds to entrepreneurial ideas around this subject, all this hoopla could result in actual progress.
Do you get the impression from the proceedings so far that the FDA is genuinely open to new ways of thinking about this? If so, that’s encouraging. I’ve been hearing the opposite view from others.
I fully agree there’s now a clear mandate for an FTC crackdown – let’s just hope they hit the right targets!
I know you know your stuff, Daniel. But I have to disagree. I’m more cautious on this.
Some of the conclusions being drawn to date are based on early adopters. These are folks with “prepared minds”, who are probably more educated on the issues, and may have an actual grasp of statistics. The general public is not ready. And I think that as a person trained in science it is my responsibility to step up and protect some of the vulnerable people from misleading claims–which is a legitimate issue still. I think we can overcome some of that with some standards. But we are not there now.
It’s not clear to me why you and I diverge so much on this. I think we probably have similar educations and access to the current genomics data.
I need to see the rest of the hearing, I only caught part of it today. Looking for the recordings as well.
That said, I have to admit I laughed out loud to hear a politician say “single nucleotide polymorphism”. I was kinda impressed, actually….
mary, the main issue i always think about is the basic understanding of health issues that the public has anyways, and the confusions that ensue from media reports on this observational study or that. in the grand scheme of things DTC genomics is small potatoes.
@razib: I understand that this isn’t the biggest issue in the grand scheme of things. But I might have said that about mis-information on vaccines in the late 1990s, assuming everyone understood the truth and the science would be clear….
But it turns out I spend an awful lot of time combatting “sciency” mis-information that people have gleaned from a web site somewhere. This is true on vaccines, GMOs, various forms of cancer woo, diet, etc. My own sister-in-law was telling me about how great her DNA diet was doing. It was everything I could do to keep my head from exploding in the kitchen. People are being taken on all sorts of health issues, and this is just the new front in that, I know. It’s still wrong.
I dread the call from a sibling saying, “Hey Mary–I just spent $500 on this test, and it says __________. Am I going to get ___________?” And at least my siblings have someone to call. And so do yours. And so do Daniel’s. Not everyone has that.
But I might have said that about mis-information on vaccines in the late 1990s, assuming everyone understood the truth and the science would be clear….
i think that analogy is weak because the mis-information on vaccines was blatantly false. the biggest problem with DTC genomics is that most people are stupid and don’t understand probabilities, nor do they bake uncertainty into the cake of any given prediction (the press doesn’t either when reporting on a medical association between X and Y; lack of replication ain’t news). the outright falsehoods (highlighted above) are i think outliers. most of the damage here doesn’t follow the power law, i think the more subtle and nuanced misinterpretations are going to be the main downside. my point is this: this is a problem with medicine today, period. so measured against that base i doubt that DTC with light regulation harms much more than what’s going on today (e.g., eating lots of refined sugar because fat is bad, going on atkins because fat doesn’t matter, etc.). on the plus side there are the minority of people who get significant benefit.
more precisely my model is this:
aggregate utility = (many people)X(slightly wrong or trivial understanding + (very few people)X(gross misinformation/falsity) + (few people)X(actionable information that yields great benefits)
so in the medical context i’d put it at a slight positive. to some extent i think a lot of the same is going on with nutrition and diet in our culture. the question for me isn’t whether DTC genomics is not as much good or bad (i think a bit good), but, whether it is good or bad in relation to the current state of medicine and law. there are other low hanging fruit out there, and DTC genomics seems to be getting slammed because it is new and easy to scare monger.
all that being said, i assume you know more about this than i do. so with my parameters thrown out there, could you correct where i’m wrong in my sequence of propositions?
thanks.
Ok–here’s my first and informal pass on your equation, with the caveat that I’ve had two glasses of riesling. I may have different thoughts in the morning:
@Razib: aggregate utility = (many people)X(slightly wrong or trivial understanding + (very few people)X(gross misinformation/falsity) + (few people)X(actionable information that yields great benefits)
@Mary: aggregate utility = (many people)X(slightly wrong or trivial understanding + (very few people)X(gross misinformation/falsity)very large megaphone + (few people)X(actionable information that yields great benefits)
or
@Mary: aggregate utility = (many people)X(slightly wrong or trivial understanding + (very few people)X(gross misinformation/falsity)selling products + (few people)X(actionable information that yields great benefits)
I have to submit without preview, so I don’t know if the HTML will carry. We’ll see….
yeah–the superscripts for “very large megaphone” and “selling products” did not work.
It’s not additive.
thanks. i got it. i’ll think on it….
While I always maintained that “there is more to the hologenome than SNP-s” the GAO report is astonishing (see hologenomics.com “news”).
Pellionisz_at_junkdna.com
Razib (and Daniel, for that matter) — Who is getting the “great benefit” from any of these DTC results? These companies are offering essentially no benefit beyond personal entertainment, with nonzero risk of misunderstandings and negative deflection of medical decision-making by consumers. I can’t for the life of me understand why Daniel is so gung-ho about these products, at least as currently constituted.
These companies are offering essentially no benefit beyond personal entertainment
i disagree that it’s essentially no benefit. some people will find large effect alleles. true, they should often have some clue because of family history, but many are not cognizant of that sort of thing. it could be that the effect is less than the cost.
Razib — I agree that discovery of a large-effect allele at CFTR or LRRK2 is potentially beneficial information. IMHO, however, presenting this information over the internet by an unregulated company leads to risks that more than outweigh the benefits, even for the affected individuals. Therefore, I stand by my claim that the benefit is essentially nil, while the risks are nonzero.
Let’s say I agree that there’s a lot we might learn from our DNA, even without doctors’ involvement. But isn’t it in all of our best interest that the industry be well-regulated?
Neuro-conservative:
The widespread use of GPUs for PC games in the 1990s has now filtered down to scientific computing. In other words, a massive entertainment market allowed the development of devices that are very useful for basic research.
If you read the PLoS Genetics paper, 23andMe really is innovating in a way that very few labs are capable of. They are running more and larger GWAS than anyone else on the planet right now. Yes, it starts with recreation, just like GPUs, but there’s no way you’re going to get rare variant signal without a massive database.
Also, I want to see peer-reviewed studies on the “risks” you and Alberto Gutierrez seem to have dreamed up. Why is the burden of evidence so selective and one-sided?
I also want the FDA to pay for the costs of compliance out of its own budget. Shouldn’t be too much, right? So they can pay for it.
> But isn’t it in all of our best interest that the industry be well-regulated?
No. The computer industry is essentially completely unregulated. Yet computers are used in mission critical applications all around the world.
How did that happen? How was safety maintained in this ostensible wild west?
Underwriters Laboratories, a private certification agency, takes care of certifying that a device performs up to snuff. There is competition among certification agencies in the area. By contrast the FDA has a monopoly on certification of medical devices and drugs, and is as dysfunctional and incompetent as one would expect a monopoly to be.
@alpha beta
How was safety maintained in this ostensible wild west?
That’s an interesting example, but not the only one. Isn’t every Western government presently bemoaning the consequences of a lack of regulation in financial markets? Five years ago credit-default-swaps were hailed as sophisticated innovations which were enabling the market to reach dizzying new heights, now they’re snake oil which destroyed the conventional economy.
On this topic more generally I have to say that it’s not clear to me how much of a disaster for DTC genetics companies FDA regulation would be. My understanding (which I’m sure will be corrected) is that DTC genetics is likely to require 501(k) approval from the FDA. A visit to the FDA website showing approvals just in June 2010 reveals some hard core medical technology (lasers, surgical equipment), but also things which seem to me to be in the same ballpark as DTC genetics (glucose tests, pregnancy tests), as well as very basic stuff (tampons, latex gloves).
Furthermore, I do think the FDA has an obligation to provide some regulation to the industry for two reasons:
1. To close down the scams saying that ‘DNA is a symptom of biology’ and that DNA can be repaired by ‘epigenetics’. I agree with Daniel that if companies like 23andme get conflated with this crap it would be a real shame. So far, however, in the media coverage I’ve seen the focus has been on the bigger, more reputable companies, so…
2. To make the companies provide a better service. I don’t buy the “lone wolf” operator theory about the tapes. Clearly these companies can’t get their employees to provide accurate information to customers. There’s a lot of possible reasons why that might be, but it seems undeniable that an across-the-board improvement is needed. I also agree with some of the criticisms of “research” and “clinical” reports from 23andme.
On the other hand, some of the GAO report criticisms (such as the one highlighted by Daniel about a person who had a heart problem but was told he was at low risk) reveal a lack of understanding of the basics of this topic by the authors. I guess we’ll have to wait and see what happens.
Back to the first comment – I hope Linda is correct and there is precedent. A few years ago there was a senate hearing on DTC, watching yesterday was like going back in time, the GAO format was so similar. Selective evidence, mistakes, falsities, loaded report (I mean just look at the title of the GAO report) etc, creating the impression that the whole industry is tainted. The judgement was damning, the FDA got knuckles rapped because they hadn’t been doing anything, etc etc. Yesterday was like a Congress remake of an old Senate movie. The point is, nothing happened regarding regulations despite the promises. That was not a good thing though, there was an opportunity and it was missed, the only thing that remained was the “fact” embedded in Google search stone that the GAO found that it was all snake oil. The only thing that is repeated from that hearing, is based on the press release that the GAO put out before the hearing even began, no notice was taken of the evidence or the defences.
This time though it should be different. The tactics of the GAO have been exposed to a wider audience, the industry, while fragile, is still more robust than a few years ago and there is more will to fight back. The best thing that they could do know is get together as soon as possible and create an industry body comprising the reputable companies, sort out the standards, and create a reasonable model of self regulation that the FDA can incorporate into it’s own oversight. Maybe the FDA would be grateful for the help and ideas, Shuren did admit yesterday that they need some innovative new regulatory models for the area.
As for utility, even if it is only entertaining, what is wrong with that? Entertainment is a strong personal utility. I believe it is more than that though and at least there is evidence of benefit (see, for just 1 example) Blaine Bettinger’s post and comments at bit.ly/4Bnysh) whereas so far I have not seen any for harm.
Just one other thing, agree with Jeff Barrett – there are definitely things wrong with the good DTC companies and they need fixing. There are things seriously wrong with other companies that are exploiting on purpose. All this makes it a great pity that the GAO did not do a proper job, a real “forensic” analysis and isolate the good from the bad, after all it would not have been difficult to do and it would gave been of great benefit.
But the agenda of the GAO I think was different – they started with the conclusions and set about to get the evidence
To those thoughtful people who feel that I should not be allowed to read my own genetic code, in case I misunderstand it, I have one question. How much are you biased by the subconscious desire to keep power (knowledge=power) in your own hands?
This power struggle gets played out in various professions all the time. Lawyers will shake their heads sadly at the idea of anyone making a will without legal assistance. They will tell you horror stories of what shocking mistakes could be made that way. Medical professionals naturally think that they know best. Look at all that training they have had! How could the patient know better? In fact for every scare story of idiotic people doing something foolish without the tender (and expensive) care of medics, it should be easy for journalists to come up with cases of iatrogenic disease. According to some sources the leading cause of death in the US is the medical system.
If the FDA closes down DTC genetic testing firms in the US, then that would be a big boost to those firms based outside the US, such as Icelandic deCODEme. There will always be a supply to fill demand. So I doubt whether the FDA would do something so foolish. 23andMe is a front runner in the field. Forcing them to relocate to Switzerland or wherever would be an own goal financially.
One particular scaremonger has spoken of his nightmare vision of someone having her ovaries removed because of a test result showing higher-than-average risk of ovarian cancer. How could anyone have her ovaries removed without surgical assistance? A surgeon has to agree.
The sad fact is that women have been scared into unnecessary surgery by fear of cancer for donkey’s years. It’s bad enough that surgeons are willing to do a double mastectomy on a woman who has been diagnosed with breast cancer. There are bound to be cases of incorrect diagnosis. One woman at least sued on later discovery that her biopsy result was a false positive. Her tumour was actually benign. Worse still is the willingness to carry out “preventative” double mastectomy on women who simply have a family history of breast cancer, without even knowing whether the woman concerned have inherited the relevant gene.
DNA testing has the potential to aid both the patient and the medical profession in making such decisions.
I have had my own experience of the GAO, not pleasant in 2006 as in 2010 – it’s here for anyone interested: bit.ly/9b3EbV
“To those thoughtful people who feel that I should not be allowed to read my own genetic code, in case I misunderstand it, I have one question. How much are you biased by the subconscious desire to keep power (knowledge=power) in your own hands?”
The issue isn’t preventing people from being able to access their own genetic information, it’s that in doing so, they shouldn’t be given misleading medical advice or information. If you want to give customers medical advice, then you have to play by the rules. If you want to accept that these gene chip assays have plenty of value with regards to intellectual curiosity but basically no clinical utility, that’s perfectly fine, and FDA wouldn’t have a dog in the fight, but if the DTC companies are explicit about that then their VC funding and customers all vanish. The money is in medical.
@Jean M: Jonathan’s reply is essentially what I would have written.
It seems to me that the regulation-free cheerleaders are trying to claim there are all these benefits from looking around. And that people can make changes from knowing the information: such as Francis Collins losing weight based on his diabetes risk which we heard at the hearing (and I heard him say on Charlie Rose as well).
In the same breath they say people won’t make bad decisions. Seriously? People will only make good decisions with information (some of it of dubious quality right now)? I doubt that.
If you want to claim people are going to make decisions based on the data, you can’t only claim the good ones. You will have to accept the bad ones too.
Jumping in late…
Firstly, Mary: as far as I know I’ve ever made the claim that DTC genetic information is likely to profoundly transform people’s health behaviour. The evidence thus far (which is admittedly inadequate) suggests that in fact genetic test data have surprisingly little effect – positive or negative – on testees’ emotional well-being or behavior. In the absence of evidence for harm, I have a natural tendency to oppose paternalistic “need to know” approaches.
Jeff – regulating a glucose test is very different from regulating a SNP chip simultaneously examining hundreds of trait-associated SNPs. It’s also very different from regulating an industry where the average life span of a new technology is 18 months, and where new associations are identified every week. Traditional regulatory approaches applied here could easily cripple the industry.
Look, my post is a little over-the-top – the tone of the hearing made me a little more dramatic than usual. I have noted in the past that I’m not simply anti-regulation; I fully support careful regulatory responses that punish the scammers without punishing innovation. Maybe that’s what we’ll see here (Dan and Linda are more optimistic than me, which is reassuring given they know a lot more than I do about the regulatory landscape). However the witch-hunting tone of the inquisition so far leaves me pessimistic.
Dan, your post was not over the top, it was right on. Some of the others are just not as informed about the FDA because they are coming late to the party.
Jeff Barrett, you are an exceptional researcher but (and I hope you don’t take this the wrong way) this is clearly an area which you are just getting up to speed on. We all benefited from your post on GWAS the other day; please consider this a similar intro to what’s going on with the FDA.
The end of VC investment in genomics
The most important high-level point to keep in mind is that 23andMe met with the FDA Commissioner in 2007 before they launched the company. Say what you will about them, they’re pretty much doing the same thing now that they did then. The only reason they are being persecuted now is because of negative press, not because of anything substantive.
Let me make this crystal clear: no investment is going to happen in US genomics startups when businesses can be randomly destroyed in less than three years after getting a personal verbal ok from the FDA commissioner for their business model!
Talk to anyone in the Valley. This has killed VC in this space for the foreseeable future. No one is going to fund a US startup in genomics. Doesn’t matter what it is if it can be declared illegal overnight. And remember, not even sequencing or software is safe, because Illumina and Knome got letters too!
It’s almost impossible to explain how bad that is from an entrepreneurial perspective. A business cannot be built in a day, it must be built with a view for the long term, and the concept of arbitrary, press-driven mid-game rules changes means that business investment just will not happen. This is what people mean when they say that investors hate “uncertainty”.
Think about this very website: GNZ was set up to review the burgeoning genomics industry. Guess that plan needs to be junked!
Basically, no venture can survive in an environment where they go from toast of the town to illegal in less than three years flat. It is reminiscent of the Soviet Union, where purges of formerly respected officials were so fast that they made your head spin.
First they came for the startups…
How would the Wellcome Trust like it if this kind of report was written on the associations which came out of WTCCC? It would be easy.
Completely unfair, right? Not least because the understanding at the outset was that WTCCC was just research and it’s ok if it didn’t completely succeed. Mid-game rules changes do not benefit anyone. Negative publicity of this kind tars all of genetics.
Well, you say “personal genomics and academic research are very different”. I agree. But then the financial markets have nothing whatsoever to do with Illumina’s calling algorithms, yet the former is being cited (even by intelligent people like yourself) as a reason for why regulation is good always and everywhere!
(Oh, and by the way, the last major financial regulation (Sarbanes-Oxley, passed in the wake of Enron) accomplished nothing but destroying the US IPO market. It did nothing to prevent this financial collapse, which is a whole separate topic.)
510(k) and PMA approval costs millions; FDA has already indicated it won’t be granted for DTC
Two things: first, it is “510(k)” approval (for medium risk devices), and second, Shuren has already said that DTC is going to require PMA approval (which is the same level of scrutiny required for pacemakers).
Thing is, 510(k) approval, costs millions of dollars, and PMA (high risk) approval costs tens of millions of dollars:
And in this case, Jeffrey Shuren and company have already indicated that they are going to set the bar so high that the approval will not be granted:
Thing is, for a company to pass 510(k) (let alone PMA) approval it needs to be set up like a bureaucracy and lobbying firm from the beginning. 21 CFR part 11 alone is no joke. You need all kinds of software just to do change tracking. Watch this three minute video to get some of idea of how highly regulated the area is.
Forget about web software or doing software updates. You just can’t, every software change has to be reported and approved to the FDA. It takes you back 20 years to shrink-wrapped software on CD which you can’t patch. If you do patch it, the FDA considers you a criminal and there are all kinds of penalties.
Any attempt to hold the DTC (or other genetics) companies feet to the fire for variablility in the risks estimated from the same data should be coupled with a similar analysis of other clinical practice. A story carried widely online examined variability in the pathology calls for small breast cancer lesions (ductal carcinoma in situ), with some women treated with chemo/radiation/surgery who may not have had disease.
The whole point is that a lot of diagnostic tools aren’t nearly as clear cut as we’d like AND that the current system doesn’t always aim for excellence (one patient profiled in that story had a diagnosis based on a read by a pathologist who was not board certified for oncology nor processed many cases). Surely the genetics predictors companies could come up with some standards, but to shine a spotlight on their issues without the proper context of other medical settings is irresponsible.
The one thing I hope DOES NOT HAPPEN is that the regulatory agencies (they work for us right?) “punt” on the issue and turns the whole consumer genetics ball of wax over to medical doctors and the medical insurance complex.
Like many, I am inspired by open-source, open-access, crowd-sourcing, bioinformatic and other open, web-based tools that allow consumers to by-pass the so-called “experts” in news media, finance, health and so many other industries that are being transformed by information technology. The economic benefits for consumers are well documented, and so, a country like the U.S. – economically sinking in a healthcare affordability crisis – might benefit (in the longer run) if it nurtured industries that helped consumers freely and openly ascertain their risks for illness without having to go through the economic choke point of an establishment of medical “experts”.
See the “health 2.0” conference, Regina Herzlinger’s “Who Killed Health Care?“, Michael E. Porter and Elizabeth Olmsted Teisberg’s “Redefining Health Care“, Andy Kessler’s “The End of Medicine: How Silicon Valley (and Naked Mice) Will Reboot Your Doctor“ and Nobel Prize-winning economist Kenneth Arrow’s classic 1963 essay “Uncertainty and the Welfare Economics of Medical Care“ for some more on this.
“If you want to claim people are going to make decisions based on the data, you can’t only claim the good ones. You will have to accept the bad ones too.”
Naturally some people will make what we may think are bad decisions. We are talking about the human species here. I see no chance of us becoming infallible any time soon, or even putting long-term health ahead of short-term pleasure in any concerted way. Anyone who is still smoking is ignoring a mountain of medical evidence. What can society do about that? Should we keep from people the medical evidence, because they won’t act on it? (Or won’t act as we think they should?) What sense does that make?
I am not against regulation of the DTC genetic testing industry, incidentally. I agree with Linda Avey that fraudulent companies should be shut down. But I do champion the principle of DTC genetic testing. My genetic code is mine. I feel that I have every right to read it. Who has more right?
@Jean M: well, we regulate tobacco, don’t we? We require warnings on packages. We monitor advertising. And we restrict access. Right? At least we do in the US, I don’t know where you are. I’ve seen even more serious stuff in some other countries.
Any attempt to hold the DTC (or other genetics) companies feet to the fire for variablility in the risks estimated from the same data
This is framing it the completely wrong way. What if people held browser vendors “feet to the fire” for variability in how they rendered HTML? Obviously, there would be zero innovation in browsers. No one could do something new because it would be “nonstandard”.
By its nature, a standardization process must follow innovation. It cannot precede innovation.
What if researchers were publicly abused for differing with their colleagues on a computation?
Or, more to the point, what if the bureaucrats federal agencies were publicly abused for saying one thing in 2007 (“23andMe has freedom to operate”) and then turning around after tens of millions of dollars has been poured into the idea to declare it illegal? Isn’t that an inconsistency which deserves “feet being held to the fire”?
Mary, giving the government more power on the basis of scare tactics and fear mongering is bad when it’s a Republican and ALSO bad when it’s a Democrat. Give them the power to regulate the behavior of consenting adults and there is no stopping them. The way government works is by using and abusing precedent to expand.
Consider this. Your company (Biological Software Testing Services, Inc.) provides services could be used to discover a drug. It is conceivable that a software error in your code could lead to a downstream mistake that endangers a life. And hence you should be publicly flogged as an irresponsible software developer should this happen, because of your failure to use design control during the development process.
Crazy, right? The problem there is the multiplication of probabilities. If the worst thing in the world is that you need to give a relative an education in probability, that doesn’t justify bankrupting an operation that is doing more and larger GWAS than any academic lab in the world.
As a concerned private citizen, I’d like to protest being called a “consumer” of my own genetic information. I don’t consume my height, weight, or blood pressure. I may consume oil, milk, or energy, but this is information about me and my life – a portion of which relates to disease.
I like to use the analogy of Gutenberg’s discovery of the printing press. Someone first looking at his converted wine press might very well ask, “Who needs all those books? No lay people can read them.” Nevertheless, we had a literature/literacy spiral, far outside the control of the priestly class. So, I think we should see personal genomics as a new medium, with similar dynamics for a literature/literacy spiral. And perhaps a similar fate for the priestly class that would intersperse itself between the medium and its interpretation.
I agree that there should be standards for doing the testing, but the process should be open. I should be able get my raw data, and interpret as I wish. If I want to use someone else’s interpretation, that is my choice, not the government’s. Just keep the process open and transparent, thank you.
I also am concerned about the tenor of the meeting – why didn’t the presenters have a chance to read the GAO report before testifying? Why were they asked to respond under oath to a report that they hadn’t read?
So, rather than lock up genomic information behind the walled gardens of special medical interests, let’s open up the research for free and open access. Let’s require that the results of ALL NIH funded research be freely and openly accessible the public.
I watched the live webcast yesterday and was frankly appalled at the biased way in which the “evidence” found by the US Government Accountability Office was presented. I would hope that the GAO will in due course release the full recordings and transcripts of all their telephone conversations. (I would hope that there is a US equivalent of the UK Freedom of Information Act and it will be possible to request this information.) It would appear that the two DTC companies in the video are Navigenics and Pathway Genomics. Both these companies now require consumers to order their tests through a doctor and/or with advice from a genetics counsellor. The GAO were seemingly unable to find any evidence that 23andMe and deCODEme, the two companies who do sell their tests direct to consumers, were misinforming their customers over the telephone. 23andMe, Navigenics and Pathway were all asked to testify at the hearing. Unlike the newspapers, they were not given the chance to review the GAO evidence beforehand but were asked to comment on these findings on the spot. In the hearing Ashley Gould the counsel for 23andMe revealed that the company actually approached the FDA for advice three years ago before they even launched the company. If the FDA were so concerned about the 23andMe test why did they not work with 23andMe at the time rather than hounding them in such a public way now?
Misleading marketing and advertising can be a problem with all companies. In the UK we have an Advertising Standards Authority and an Office of Fair Trading which oversee such issues. Is there no such equivalent in the US? There are indeed fraudulent companies which need regulating and investigating. I find it incomprehensible that the Congressmen and the FDA are focusing all their attentions on the responsible companies like 23andMe while seemingly ignoring the fraudulent scam companies such as the supplement companies shown in the video. Why were representatives of these other companies not called to give evidence?
Genetic tests are predictive not diagnostic. The average consumer is only going to test with one company and simply could not afford to test with multiple companies so the different reporting methods and different SNPs tested are unlikely to be such a big issue. A patient could just as easily consult a number of different doctors and be given different diagnoses and prescribed different drugs. A predisposition to breast cancer could easily be uncovered by researching your family history. Perhaps the FDA should also be regulating family history research! Whatever decision is made it should based on evidence not anecdotes and not scaremongering. The Congress team admitted that they hadn’t a clue how many people had taken genetic tests and they had not done any market research. They were completely ignorant of international initiatives such as the Human Genetics Commission’s Common Framework of Principles for DTC companies which is due to be published shortly after a lengthy two-year consultation process. Over 50,000 people have reportedly been tested at 23andMe. If these tests are such a big problem and uneducated consumers are indeed making decisions which are adverse to their health based on their test results how come the FDA, Congress and Rob Stein of the Washington Post (who actively sought negative experiences from DTC consumers) have not been able to find any examples?
“They (23adnme) are running more and larger GWAS than anyone else on the planet right now”
I disagree. Their sample cannot possible exceed more than 100,000. I don’t know what Gould was talking about running GWAS day and night, but really you need to examine quality and not quantity. What do you think the breadth and accuracy of the phenotypes they are a evaluating is? For example, do you believe they have accurate lipid fraction measurements on all of their customers in order to do GWAS on cholesterol metabolism? No. They have a database that we can learn from, and they may be good at validating markers, but dont give them too much credit. What they could do, that would be powerful if they could follow their cohort long enough, is to begin to validate penetrance risk estimates of disease associated with markers they are looking at.
@Debbie:
(hope that code works…if not, that’s a quote…)
This is one of the things I don’t think is a fair point right now. Ok–50k people have been tested. As I mentioned in my comment I consider these people “prepared minds” and early adopters. They are not the average person. That’s a self-selected group who may not be average. They may have better access to information. They may have better critical thinking training. They may have better science/math/medical backgrounds.
And we don’t know if there are problems. Absence of data is not the same thing as absence of issues. How would we know if there were bad choices made subsequently? Has it raised people’s anxiety levels? Has it caused family tensions? Is it a representative cross-section of people? Where has that been surveyed? We don’t know.
Mr. Yarrow,
While I agree that the way the FDA has gotten into this has been less than stellar, I think using words like “persecute” and phrases like “It is reminiscent of the Soviet Union” undercut your points. Why should these companies not have to follow standards that others do if they are providing medical information? If you want to argue what “is” is, go ahead, but you are going to have a hard time convincing me that abacavir hypersentivity status is entertaining.
While I agree about the arbitrary timing of the FDA actions, you underestimate the changes 23andme and others have undergone since they began. 23and me and others have added more conditions and increasingly pushed the envelope of their medical language in marketing. They were much more conservative at the outset. Again, Spit parties were not being held to find out your abacavir hypersensitivity. That just aint fun.
You say “A business cannot be built in a day, it must be built with a view for the long term” – any VC company who thinks that the current DTC model is a viable long term model is deluding themselves. Complete raw sequence data is going to be available to everyone before long. Providing sequence to people will be a non-starter before long. The companies that figure out how to constantly and accurately annotate the data for individuals and their providers are the ones that have a chance to persevere. Clearly the companies we have are not there yet – the differences in risk predictions are not accurate enough (and as an aside, that doesn’t mean they have no utility, I believe strongly in the personal utility of this data, even absent clinical utility, as does the general public). Maybe a lasting model can come from 23 or Navi, but even given the best of times and no regulation, the current models are transient at best.
Nothing has been declared illegal overnight.
Illumina is not going out of business.
I think Linda is right that after the grandstanding and dog and pony show the real stakeholders will get down to business to see where things need to go. FDA had its moment in the sun, now the hard work needs to begin. I believe FDA has little or no idea exactly how they WILL proceed – but you have got to start the bidding war high if you want to get the companies to negotiate. PMA and 510k and the GAO report are tools to get things rolling.
Ask yourself, did the risks FDA is citing emerge overnight? No. The FDA needed a way to get a serious ball rolling and they grabbed a couple of big mallets and started swinging. It seems like it is rolling now.
“What if people held browser vendors “feet to the fire” for variability in how they rendered HTML? ”
– seriously?
You’re making a LOT of assumptions in the perceived “reputability” of the 4 companies you seem to favor, and your sharp demarcation of “bottom-feeders.” (And even companies with ‘reputable’ science these days are often compelled to engage in unethical/disreputable marketing.)
Mary
No has the answer to these questions because no one has done the research to find out who these people are who are taking these tests. With 50,000 people in their database 23andMe are going to have a broad spectrum of testees and not just the well informed early adopters. Many of our 7000+ ISOGG members who are primarily genealogists have tested with 23andMe. Many of them took advantage of the 23andMe sale at the end of last year and the recent $99 DNA day sale. Some of them are indeed struggling to understand centiMorgans and the significance of the size of their matching haploblocks in the Relative Finder. However, we’ve not had a single report of anyone misusing the medical predictions and making incorrect decisions. On the contrary, people have been empowered by the information and been inspired to make changes in their lifestyle. There is still no evidence that any of these tests have caused any harm. On the other hand there are thousands of gun deaths each year in the United States. Why does the FDA not regulate guns?
And we don’t know if there are problems. Absence of data is not the same thing as absence of issues. How would we know if there were bad choices made subsequently? Has it raised people’s anxiety levels? Has it caused family tensions? Is it a representative cross-section of people? Where has that been surveyed? We don’t know.
Sorry, The last paragraph of my published comment was part of Mary’s reply and should not have been included.
Great discussion. I share Daniel’s disappointment and the anger of many others. I think the companies have issues they need to address, but as a taxpayer and close observer of this industry I resent the need to put on a congressional reality show that appeals to the lowest common denominator.
23andMe Video recordings from the forum “Genomics and the Consumer: The Present and Future of Personalized Medicine” Hosted by California State Senator Alex Padilla and 23andMe Wednesday, July 14, 2010 are now available on their site: https://www.23andme.com/policyforum/
Looks as though 23andMe has actually been pressing for some kind of regulation.
Also a fascinating presentation from Leroy Hood, MD, Institute for Systems Biology: P4 Medicine (Predictive, Personalized, Preventive, & Participatory): Catalyzing a Revolution from Reactive to Proactive Medicine
@Debbie:
Um, guns are pretty heavily regulated. Require licenses, operator safety training. Warning signs at point of purchase. Stuff like that. It would be ok with me if the FDA regulated them too, but as I understand they are not considered medical devices under any current definition.
Do you think it’s possible people tell you the good stories and keep bad stories and anxieties to themselves?
That said, I do have to say how interesting I think it is to see everyone’s support for industrial science. That’s not very common. And after the recent PepsiCo drama at ScienceBlogs I wouldn’t have predicted the strength of the support for industry.
This is a great discussion, but the argument that regulation of DTC genetics will somehow limit consumers’ access to their genetic code is a strawman. The FDA regulates plenty of over-the-counter products that consumers purchase every day.
No one is saying that patients should not have access to this information, just that the information they receive be accurate. What’s wrong with that?
Mary
On the contrary, on our mailing lists we tend to get the problem stories and the people who can’t understand their results. We were also amongst the first to know of the 23andMe sample mix-up because the initial reports of strange results came through on our lists. Because we were all able to work together as a coummunity the problem was quickly identified and reported to 23andMe. Unfortunately these same people have been hounded by reporters from supposedly reputable American newspapers who have twisted and distorted their comments to put DTC testing in an unfavourable light. They have found the trauma of dealing with the press far more upsetting than the wrong results they initially received.
23andMe responds: GAO Studies Science Non-Scientifically
http://spittoon.23andme.com/2010/07/23/gao-studies-science-non-scientifically/
There are two subjects that one should not forget that are operating here.
Government 101
‘Never have an inquiry unless you know the outcome’.
Media 101
‘Never let the facts get in the way of a good story’.
Government 1, Media 1, DNA-Sequencing technologies 0.
All quiet a sad day for Democracy and the freedom to invent.
The analogy to guttenberg seems pretty compelling to me from an education and innovation perspective.
From a rights perspective, knowing your own genome seems pretty inviolable.
Individuals should be free to get their genome sequenced/genotyped, companies should be able to charge to provide that information and interpret the non medical aspects e.g. Ancestry, with light touch regulation or self-regulation ensuring analytical validity (although the Market will do this as well), recognising that there can be a grey area between the medical and nonmedical aspects. Drawing a line does not appear too tricky to me. The grey area is occupied by odds ratios equivalent to looking in the mirror.
Caveat emptor
If companies want to make money from interpreting the medical aspects of a genome beyond some recreational odds ratio. They need to demonstrate competence in assessing clinical validity and communicating results in a responsible way. We can argue about what that means, taking current clinical genetic practice ( which itself is evolving) as one model that achieves this. Codifying these requirements for companies will likely have a positive effect on clinical genetic practice, which is not quite as rigorous or robust as it could be. Given the potential harms it is reasonable for government and the medical profession to be involved in establishing this regulatory framework.
People will get their genomes and use free open source tools to interpret their disease risk themselves. This is the logical consequence of allowing individuals access to their genome, and having research results open access. If they find anything scary the sensible ones will go to see their doctors – just like finding a lump. People find lumps and get bad medical news before their doctor knows every day. People find lumps and do not got to their doctors every day. We do not stop people from assessing the lumpiness of their own body.
Coping with this inevitable new route for concerned patients seeking medical genetic advice, presents a challenge to the medical profession, meeting that challenge will be to the benefit of doctors and patients.
As an addendum, who else is delighted that triplet repeat disorders are so tricky to incorporate into genotype chips and short read sequence analyses, nicely preventing some specific potential harms from this first stage of opening access to our genomes.
dave, should you be interested in a reasonable discussion, my attempt at polite answers is below.
They have 50,000+ people (more than WTCCC1) and more phenotypes than any other project by far. Moreover, they can actually do iterative GWAS where they do an association and roll out more questionnaires. And if unimpeded by government, at their growth rate, in a year they would easily be the largest without question. Remember, they are in the black on every sample while an academic research project loses money.
To explain in simple terms: every night, they re-run their regressions with an update step (similar to the Kalman filter) to see whether new data has led to new associations. In general, as sample size increases, you should see nonzero coefficients come off the axis and have a stochastic walk towards their final, stable equilibrium values (as sample size approaches infinity).
This is not a useful criticism, as it could equally be applied to the massive case/control consortium studies (like WTCCC, I believe) which used a binary phenotype for diabetes rather than a lipid fraction endophenotype. Sometimes the perfect is the enemy of the good. Better to get some data via web questionnaires (including the kinds of fun things that NIH won’t fund) than to get none at all by asking for expensive phenotypes which require physical presence.
Give credit where credit is due. 23andMe built something pretty cool that more than 50,000 people around the world actually paid cash money for. Their ancestry painting and UI is awesome and it’s really not that easy to work with that much data. In terms of impact on the world, 23 is a lot more important than the vast majority of forgotten grad student theses. Look at Google Scholar and hope that something one invents gets that many citations.
Well, I agree that following their cohort for a long time is a great idea, but in case you didn’t notice, the government is trying to put them out of business. It has already caused serious damage. It’s very hard to recruit or to raise money when an unelected bureaucrat’s arbitrary decision could end the company overnight. Remember, this is a company run by scientists like you and me, like the people down the hall with pipettes. They just poured years of their life into this only to be slandered and attacked in what can only be called a show trial.
@Frances:
Ding! Ding! Ding! And that their information doesn’t get misused. How can people be opposed to that?
I can’t understand the melodrama, really. Help! Help! I’m being repressed! This is a major over-reaction to the hearing. I’ll wager $10 that they don’t kill this industry.
The year 2010, with the most devastating financial times since the great depression may best be remembered by the sheer idiocy and waste of money by the US Government. I am referring to SNP-Gate. To see the US Government have an entire committee of people launch this type of witch-hunt is frightening . To find only one person is a geneticist on this committee and they do not agree with this restriction of the information is even worse.. Starting with the initial issue of a letter pertaining to regulatory compliance on a “medical device”, to the entry in the report one donor who had a pacemaker implanted to treat an irregular heartbeat was told that he was at decreased risk for developing such a condition , is clear there is nothing of any real merit here and no real knowledge of what genetics can mean to the average person All that remains is a yet to be exposed backroom puppeteer attempting to take control of our personal liberty and ability to know our health risks and personal medical information. Soon we will be told what to say, think and even feel according to the highest bidder to pay someone or one’s overblown salaries.
Fortunately I am in the East now that I don’t need to put up with this — yet —
By the way, how come I can only see two out of the possible 52+ comments???
Mary, again, I will predict that you have just come to this party and have not been following it for months. Considering this quote:
But this is the heart of the matter. Go ask the FDA, point blank, whether you can start a company that offers raw, unannotated genome sequence directly to individuals.
Try it. Seriously. Email Hamburg, Shuren, Gutierrez, Mansfield, Harper. Email any and all of them.
You will not get a straight answer.
We know because Biggers tried it during the hearing.
Again. Try the experiment yourself. Their emails are public. Just for a moment, stop trying to convince us and yourself that this is not happening. I know why you’re trying to do it, because you seem like a moral person. If something bad was happening, you’d have to commit yourself to stopping it. Hence you need to believe, to convince yourself, to hope against hope, that this is actually all ok and somehow not really so bad.
But again. For just a moment, instead of arguing fruitlessly with people on a blog, send an email asking the FDA officials point blank whether individuals may sequence their own genomes.
If by a miracle you do get a direct answer, it will be a boon to us all, because no one knows what is and what is not legal. The FDA is just making it up as it goes along, sending out letters willy nilly. There is nothing you can google, no text on the FDA website that lays out exactly what they object to. And this of course raises the question of why the people who are taking great risks to innovate by founding startups are being lawlessly harassed by government.
Seriously, you don’t understand. The DTC testing industry is ALREADY DEAD. In the Wall Street Journal, Shuren declared DTC subject to PMA approval, which costs tens of million of dollars! People are quitting the companies by the droves. 23andMe’s former director of regulatory affairs left for NextBio. VCs have refused to re-up. There will be no Series X+1 financing for an industry with no growth potential.
Non DTC companies are by no means dead, but every single one of them is worried and uncertain about where and when the FDA will strike. Remember, Illumina got a letter, so no one is safe. And Illumina got this letter for selling RUO products, which are used by every single clinical lab.
http://lists.utsouthwestern.edu/pipermail/histonet/2006-December/026469.html
Now, lest you think the hapless pathologists are bad guys, MANY reagents and instruments are only available RUO because it is prohibitively expensive (remember, millions or tens of millions) to take them through the bureaucracy of 510k or PMA.
Yes, that includes the GA-IIx and the Hi-Seq. RUO all the way.
In short, a letter like the one to Illumina is a threat that the FDA may attack your business with civil and criminal penalties unless you comply. Comply with what, you may rightly ask. Well, they won’t tell you, partly because they don’t know themselves, and partly because they want you to just absolutely go limp and confess every “sin” .
Predicting what they will is like reading the Kremlin. There are massive internal politics at such a large government agency. Yet your operation depends crucially upon the picayune decisions of a remote bureaucrat who is influenced (a) by the press, (b) by the whims of a politician, and (c) by a desire to advance in status. S/he is not influenced by citizens’ petitions, or by the facts, or by a simple sense of right and wrong. They just follow orders.
The only realistic form of resistance left for DTC is if Anne Wojcicki decides to use her billions to wage a lawsuit against the FDA for the right of individual access to genome sequences.
And she might well win. The FDA has been found guilty of abuses in court many times before, most recently by a small company (Utah Medical) which subpoena’d internal emails in which FDA officials confessed to “massaging” the FDA-483 issued to Utah Medical. The looming threat of a lawsuit powered by Sergey’s billions is thus perhaps the number one thing that gives the FDA pause, because they know they could get pummeled in discovery.
Just as an indication of the GAO methods this is interesting. They have made a lot about the breast cancer “diagnosis” issue but in their report the account is incorrect. On page 16:
“You’d be in the high risk of pretty much getting it,” is how a representative responded when our fictitious consumer asked if results indicating she was at above average risk for breast cancer meant she’s definitely getting the disease”.
On their video the actual question is “So if I am high risk does that mean I will definitely get breast cancer?”
She indicated that she was at high risk, not “above average risk” and the reply, although clumsy, was correct!
“The only realistic form of resistance left for DTC is if Anne Wojcicki decides to use her billions to wage a lawsuit against the FDA for the right of individual access to genome sequences.”
It would be quicker and cheaper to draw President Obama’s attention to the issue. He is solidly in favour of reform of the US medical system along the lines outlined by Leroy Hood in his presentation on P4 Medicine (Predictive, Personalized, Preventive, and Participatory).
I think Steve Y is correct. By ignoring this industry for years, CLIA, FDA, et.al. Laid the ground work for scientists and computer scientists to pretend they were physicians. But remember, this NEVER would have happenned had 23andMe et.al. Stayed away from PGx and BRCA testing. The, “it’s not really clinically important” argument might have saved DTCG. But now it is crystal clear, the personalized medicine movement is medicine and will be regulated as such and the only people investing will be huge companies with deep pockets. I wish rather than ridicule my blog and comments for 3 years, some of these people would har engaged me to help save them from this. But now it is too late. Maybe clinical LDTs may want to follow my advice and amputate the gangrenous limb before sepsis sets in and the whole organism dies.
Steve
@Steven Yarrow:
I still think there should be QC and standards around this for accuracy, and regulations to prevent misuse. I’m sorry you are opposed to that–we’ll probably never agree on this then. I have talked to a guy who wants to offer BRCA1 testing to his neighbors from his basement. I think that’s wrong and irresponsible, because there’s no way to evaluate his sequence quality and data security.
Even publications in Science are finding “equipment defects” leading to bad data: but at least we know.
You faulty conclusions on my presence in this arena make me question some of your other conclusions, such as “The DTC testing industry is ALREADY DEAD.” But it’s hard to see exactly what you are saying between the hyperventilating such as “they want you to just absolutely go limp and confess every “sin” .”
But at least it’s not flat-out conspiracy theory like that other commenter. Oy.
> I still think there should be QC and standards around this for accuracy
Of course there should be QC standards. Where you and I disagree is that you believe that the FDA is qualified to impose them.
After all, who knows more about sequencing? The FDA or the people inventing sequencing machines?
And who knows more about genetic risk factors? The FDA or Kari Stefansson of deCODE, the MD/PhD who discovered them in the first place?
Top graduates in academia with good publications will either do startups or head to a good academic job. Only the bottom of the barrel ends up in government.
It is thus your presumption of competence which is wholly in error. These are not philosopher kings, they are cops on a power trip. Again, I challenge you to email the FDA, ask them point blank whether you can offer someone a personal genome sequence without being thrown in jail, and post the response.
> Even publications in Science are finding “equipment defects” leading to bad data: but at least we know.
Again, this is a great example. It wasn’t an FDA official who figured that out but rather the scientific community, including people from 23andMe and deCODE.
http://www.newsweek.com/2010/07/07/the-little-flaw-in-the-longevity-gene-study-that-could-be-a-big-problem.html
There’s already a QA system built into science: it’s called peer review.
Great discussion! If genetic testing is going to be available to the general public then there should be some sort of regulation for quality assurance, like the number of variants that the company analyzes or the certification of the labs used. There should be some standard. I don’t think this is be-all and end-all of personal genomics. If we look at health care in the long term, then personalized health care is the only way to improve diagnoses and treatment. And the ONLY way to have personalized health care is use his/her DNA information to predict and prevent the disease from developing. The tragedy in all of this is that predictive medicine should have been the first step in introducing genetic testing and its ability to assess risks. DTC genetic testing should come later.
Even Peer Review can fail. Remember, the longevity study was
A. Published
B. Promoted by Major Media outlets
C. Planned to be offered as a test in short time
We all need to sit back and take a few years to let this all pan out prior to hyping it in am effort to:
1. Build a market that doesn’t exist
2. Generate interest in a VC round to get your investment out 4Mil Anne W(ahem)
3. Create a “revolucion!”
That type of hype and lack of QC in medicine is inevitably dangerous to public and patients.
> There should be some standard.
There is. It’s called CLIA.
Dr. Murphy,
You’re either hypocrite or clueless – your example of BRCA testing is part of the problem, not the solution to the healthcare in the US. Most patients do not benefit from BRCA1/2 testing. Currently the mutations in these two genes are not prognostic. Studies have shown that breast cancer patients with BRCA1/2 mutations who had prophylactic double-mastectomy had the same breast cancer specific survival, distant disease free survival and overall survival curves as patients that had breast-conserving surgery (re EBCC in Barcelona last March). While I’m not endorsing GeneWize (the FTC should shut them down), at least the $140 bag of “nutrients” it sold the GAO investigator would not cause harm. I can’t say the same about the clinicians who want to keep the science hostage and send patients in drove to Myriad.
@George
“Most patients do not benefit from BRCA1/2 testing.”
I normally wouldn’t even respond to that comment.
But you insulted me, so I will begin.
1. You made no mention of prophylactic oophorectomy, so I assume you don’t know that it virtually prevents 95% of Ovarian Cancer. So anything else you say is either ignorant or pejorative.
2. I don’t like you doing the same thing the pro DTC people are complaining about the GAO doing.
3. It is people like you trying to play armchair doctor ( if you are an MD, stay away from my Hereditary Cancer patients please)
Your study is one, your comments probably won’t be supported in this thread.
So go back to school. BRCA is more than breast cancer buddy.
@Steven Yarrow
CLIA certification (as well as ISO accreditation) is for lab protocols and standards but it doesn’t certify/validate the tests or its analysis that’s being provided to the consumer.
Dr. Murphy,
If only the human body was that simple … I’m sure you know about Noah Kauff of MSK and his colleagues recent study of risk-reduction salpingo-oophorectomy in BRCA1/2 carriers … and that most research related to BRCA1/2 has been done on large families with many individuals affected by cancer, which mean that individual risk estimate may not accurately reflect the level of risk for carriers in the general population … and that breast cancer patients are at higher risk of gynecologic cancer regardless of BRCA1/2 mutation status. I’m not here to debate with you who read more scientific papers or attended more medical meetings. I’m not that young anymore.
You’re right, BRCA1/2 mutations are associated with many other cancers – colorectal, gallbladder, melanoma, pancreatic, just to name few. This is precisely the point. We know so little about that, that we all should embrace anyone willing to advance our knowledge and access to information.
The Church in Europe tried during the middle (dark) ages to limit access to knowledge. We all know how history judged the Church for doing that. Or maybe the lesson is that it is human nature, and every generation has its own group of elitists who claim entitlement to knowledge.
Still haven’t gotten a good answer to this from anyone.
Mary, can you clarify your thinking on why I need approval from some third party to look at my own body and my own genes?
The FDA blocked women from accessing Plan B in 2004. This was a direct assault on women’s health and a shocking violation of our rights. It was a reminder that you cannot trust a government organization to act in progressive interests.
If you give the FDA the awesome power to stop us from looking at our bodies, our daughters will live to regret it. A Democrat is in power now, but if a Republican is in the White House, who knows what they could do? I do not want appointees of the Romney or Palin administration poking around my genome.
Steve: “But remember, this NEVER would have happened had 23andMe et.al. stayed away from PGx and BRCA testing.”
Every citizen has the right to know his/her bases at medically relevant SNPs. My health is my own responsibility, so why does the FDA and physicians as gate-keepers want to be involved?
The FDA / USA government cannot restrict 23andMe/deCODEme/etc. from
a) analyzing our genome and providing us with our raw data, &
b) providing us with links to published scientific studies, which were often funded by the USA government itself via the NIH (National Institutes of Health), reporting “base … at SNP … is associated with a … higher/lower risk for disease …”. Direct-to-Citizen genetic testing companies also have the right to repost these results in summarizing paragraphs, since access to the literature is controlled by high charges of the publishing houses.
Nhien: “CLIA certification (as well as ISO accreditation) is for lab protocols and standards but it doesn’t certify/validate the tests or its analysis that’s being provided to the consumer.”
The requirement for validation of a test system under CLIA is 21 CFR 493.1253.
George,
That Middle Ages talking point is so 2008. It advances no discussion here and is not applicable. Why? No one is restricting access to you going out and buying your own sequencer and looking at your own genes. Nor is anyone restricting you testing your friends FOR FREE.
Further, you can join a research project like the Coriell Personalized Medicine Collaborative or the Personal Genome Projects run by Dr Christman and Dr Church respectively (I just got the tremendous irony of their names BTW)
What the FDA and congress are regulating is the SALE of human genome samples with interpretation.
When that interpretation falls in line with medical diagnosis, the regulation is especially important as the operative primum non nocere (how’s that for Church language?) then applies.
As for the applicability of screening the popoulation AS A WHOLE for BRCA1/2. I agree, there are many more genes involved in the penetrance etc. This is PRECISELY WHY WE NEED TRAINED PROFESSIONALS TO IDENTIFY HIGH RISK FAMILIES!
Dr Offit (Director Cancer Genetics MSKCC) and I sit on the CPMC ICOB together and we do talk about these things. I, more than anyone else have been quoted as saying Mendel is an idiot….Modifier Genes, Epigenetics, I get it dude…..The human body is complicated.
Which is why going around saying “You are in the high risk of getting it” sounded so scary when heard on YouTube……
Let’s face it. How SHOULD have this been done?
With Doctors like Offit, Khoury, Evans, in a clinical trial.
Then that should have been moved to the market.
I think that is what Dr. Christman and Dr Church are trying to do.
So, instead of “Jump the freaking Shark” like Google/MDV/KP/Serge/etc tried to do, let’s wait to get some reasonable scientific data……
That is why we have laws and regulatory bodies. Either you respect that or you don’t. I will never change your mind about that. But hell George, go by yourself a sequencer and have at your liberation. Just don’t try and sell it in the American Economy. Try England or Asia.
@Dirk,
You are right. How could I have been so stupid? You can go buy your scalpels at McKesson.
But seriously, doctors forever have fought the whole, cutting is important, cognitive medicine is not for ages. This is reflected in our fee schedules handed down from the mount by CMS.
Just like the right to (name your item not stated in the declaration of independence) It does not mean the government or the corporations have to be forced to provide you ALL THE MEANS needed to obtain that.
So go buy yourself a sequencer or hang with your buddy George and you can go hack your health! No one is stopping that!
What they are going to be stopping is
1. Companies trying to play doctor and make false claims
2. The sale of a human diagnostic without proper regulatory approval
So stop going so crazy here. You want your genome, go get it. Didn’t you see how to isolate DNA with some simple home supplies on Bill Nye?
All that Congress and the FDA are now doing is finally enforcing laws that already exist. If you don’t like those laws, go vote for tea part candidates and libertarians who want to remove laws. We saw precisely how well that worked for medicine in the, I don’t know BC to 1900 AD.
@Jane:
Jane, as this is not what I said, I don’t know how you want me to clarify it. You can look around all you want. Buy a PCR machine on eBay http://www.boston.com/news/science/articles/2009/05/11/do_it_yourself_genetic_sleuthing/ Go nuts.
However, I do think that if you are buying a product from a provider, that provider should be responsible to have some quality standards for the information, and should protect you from the mis-use of your genome. I think you should have privacy rights around this data. And I think that false claims made on genome-woo by people selling products should be illegal. I’m sorry if you disagree. But I hope this clarifies my position.
It’s kind of how I think about the new consumer protections on financial stuff–I’m for that too. We’ve seen what happens when an industry self-regulates for too long. The free market is not always the best source of consumer protection.
This is a pretty disappointing “conversation.” Are there any non-Godwin arguments available in defense of DTC?
I’m not surprised to see a “Neuro-conservative” come out in favor of untrammeled state power.
Don’t want a genome scan? Don’t get one. But stay the HELL away from my body.
QED
Neurocon, not sure where the Godwin is here. Can you go into detail about what changed to make Times invention of the year 2008 into a danger to life and limb? I’m really not persuaded that 23andMe is a crime against humanity that deserves a congressional investigation. Seems like harmless fun. Dont we have more important things to worry about?
@Hannah
“Can you go into detail about what changed to make Times invention of the year 2008 into a danger to life and limb?”
Can you explain to me what made DTCG Time’s invention of the year in 2008.
I mean, other than millions of PR dollars to hype a “fun” test.
Seriously, 2008 Time’s invention of the year, 2 years later <65k sold?
Meh…….
The reason for the proper enforcement of the law are the new risks posed when medical diagnostics are not regulated and then sold at Walgreens or via Amazon.
I am going to have to come down on the pro-genetics side of the movement here. I have read all your comments Dr. Murphy and still must disagree. What I tend to find frustration in this debate is that the GAO report heavily criticized these companies for providing information that is of little clinical use. Yet at the same time, including screens for BRCA and APOE alleles that have well documented clinical utility makes them diagnostics. It really is a dammed if you do dammed if you don’t kind of situation.
The other hypocritical aspect of current FDA policy is their relationship and regard for dietary supplements. The GAO report blasts the use of genetics with customized supplement regimes. Though instead of going after the supplement manufacturers claimed that the geneticists are the bad players here. Look at OSR, it was an industrial chelator labeled as a supplement that was on the market for years and it only recently got enough pressure to be pulled. Though the creator was still able to do “a last week of availability advertisement”, milk the situation, and liquidate as much of it as he could. Currently, I could legally go into my backyard and randomly pick weeds based on divination, put em all in a blender, and sell it as a “heart healthy” supplement and that would be perfectly alright. As long as it doesn’t specifically say it is for an exact disease I can spin all the “cardiac wellness” I want out of it. That is safe. On the other hand, a company using current research and legitimate science can’t try to work out my cardiac disease associations, that is unsafe and way to risky for consumers. This is politics not evidence based medicine.
Yeah, it would be great if we could perfectly integrate these into existing clinical practices though as it stands we are still doing $5,000 dollar PCRs. I was on my neurology rotation and we spent almost $20,000 dollars on the genetic work-up for a gentleman who had dementia in his 30s. I know cause I ended up helping him through getting reimbursements from his insurance. I understand why they end up being so expensive and I am not making the same “Curse you profiteering doctors!” arguments as per above. Getting clinical valid diagnostics tests has a lot more overhead besides just PCR primers, though our system pays dearly for a perhaps marginal increase in validity.
The argument about being able to go buy a PCR machine is also a pretty big straw man. We are pushing the realm of the $10,000 whole genome. Those kinds of sequencers won’t be in the hands of the hobbyist anytime soon. However, it wouldn’t take much for 23andMe to switch from using SNPs for associations to directly inputting that into their existing and quite elaborate and very educational interface. How long do you think it will take for the medical profession to develop a similar UI for primary care physicians to use on entire genomes? I don’t know where you are but at the University of Colorado Hospital where I am doing my training we still use paper charts.
Maybe a better position would be that I am for both. A loosely reigned public option offering rapidly expanding and easily affordable screens and a well-tuned medical option that can precisely follow up on test results.
@Hannah
23andMe don’t claim that their test are used for diagnosis. So does it follow that they shouldn’t be as regulated as a company like Existence Genetics?
Nhien, thats a good point. My scale is used to measure my weight, and my treadmill measures my heart rate. When I go to my doctor she will take my weight and heart rate with medical grade instruments.
In other words, just because 23andMe reports variants doesnt mean they need to be held to the standards of a medical grade test.
In the US we have largely been able to avoid this on stem cells, I hope we can avoid it in genomics as well: http://n.pr/ajlDUW
Anyone heard of a story called The Emperor’s New Clothes?
There is a lot of hysteria in the original posting and much of the response. Whatever the flaws in the GAO report, and it looks pretty solid to me, the fact is that even the big players who have enjoyed demonstrating their ethical credentials have, on a number of fronts, been caught seriously short.
The GAO report used an undercover operation to illustrate a number of points that many people have been making for a long time: the science is too immature for clinical application, the disparity between what you have to claim to make a sale and what is justified by the clinical evidence base is rather large, family history is a better predictor of heritable risk than what we know about genes; the assymetries of information between consumer and company are huge.
As for the complaint that the report was one-sided and that the last week has been a stage-managed PR exercise – well I say fight fire with fire. After all, we are talking, in one case, about a company who, at no small expense, jetted into Davos to give away their tests to the world’s political and economic elite.
McArthur talks about the difference between the reputable and disreputable players in this industry. In the course of six years of research in this area I have had a chance to talk to many of the big companies and quite a few of the small ones, firms that have been around for over a decade and some that are just a few years old. Each company says the same thing: their products and services are ethical, useful and safe, but there are some real cowboys in the market and we need regulation to sort the good from the bad.
Regarding regulatory impact – any VC company who invested in this area not expecting FDA involvement at some point (i.e. once there was a Democrat in the White House) deserves every dollar they might lose on such a misinformed gamble. As is so often the case in the current discussion, too many people are assuming that regulation will mean over-regulation, an assumption with no base in credible fact. Anyone wanting a further taste of my views can go to:
http://content.karger.com/ProdukteDB/produkte.asp?Aktion=ShowFulltext&ArtikelNr=313330&Ausgabe=254333&ProduktNr=224224
I think the last time a bloated, hypocritical, self-interested entity tried to ban the people from reading a very important book because the entity alone claimed the ability to interpret the contents, there was a revolution. The entity was the Catholic Church, the book was the Bible and the agent of its discomfiture was the Reformation.
Having driven many Americans to foreign countries to buy drugs, made regular doctor visits unaffordable and continuing in their cynical attempts to prohibit choice and the general availability of information, the AMA/FDA complex looks panicked. Goody! I hope the 23 and Me founders, who are truly innovators and experts, become much richer. I know I will continue to buy their product. And by the way, I won’t be seeing an MD from the medico-governmental complex to find out what it means – I’ll consult someone who actually knows what they are talking about – like the PhDs at the reputable companies.
Shorter Stuart Hogarth: Anything I don’t like should be illegal, oh and by the way cite my paper.
Both Dan B and mrmr make classic mistakes which I highlight in my article:
1) conflating regulation with proscription – regulating something is not the same as banning it; it is about setting and enforcing standards, something the industry claims to want
2) Dan B equates unmediated access to the genome with access to the Bible. Once the Bible was translated into people’s own languages they could read it and make their own interpretation. But as Dan himself acknowledges we cannot do that with our genomes, we need expert assistance, so the only question is who will be the experts giving that assistance – surveys consistently show that the public mostly want doctors involved.
Furthermore, and to return to point one, any expert, including the ones at DTC companies, in effect becomes a gatekeeper ie someone who takes on a regulatory role, setting and enforcing standards (that is why companies differ in what SNPs they report, which conditions they report on, whether they offer telephone counselling etc). The GAO report does a very good job of demonstrating the dangers of leaving the regulatory role entirely in the hands of industry. Again, just to remind you, the leading companies have all stated they welcome Federal input – well now we are going to get it.
Hi all,
Just a quick note to say that many of the Genomes Unzipped authors were away over the weekend (including several of us on the same trip), and thus weren’t able to engage in this excellent comments thread as it developed. We’ll hopefully be putting up our thoughts in a separate post this evening or tomorrow morning, and chiming in here when possible.
Also, a quick point about commenting etiquette here: steer clear of profanity (since we don’t want this page blocked by high school obscenity filters), watch the ad hominem arguments, stay on-topic and your comment will survive. That’s not too much to ask.
The hearing, no matter how much it hurts, can only be good for the long-term future of this nascent industry. The sooner it is properly regulated the sooner it will gain credibility.
Here’s a solution to the negative press DTC is receiving – get Rupert Murdoch and the rest of his buddies to invest, that’ll shut ’em up!
Hogarth, you do not understand how the FDA works.
By setting the bar for DTC companies at the PMA level, they are intentionally and effectively proscribing them. PMA is a long dark tunnel that costs $50-$150 million to get through, as cited above in this thread, and that’s *if* the agency wants to give you a chance. With DTC they do not want to give a chance. So we are talking about banning DTC, full stop.
These companies don’t have $50+ million lying around. This kind of thing effectively puts them out of business.
As for whether VCs should have anticipated that Obama would screw them, well, Obama serenaded them during the campaign didn’t he? No one elected Obama to declare war on genetics.
Steven Yarrow,
>”Top graduates in academia with good publications will either do startups or head to a good academic job. Only the bottom of the barrel ends up in government.”
How do you determine who is a “top graduate” in the US? “Startups” reward those with the most effective business skills. Academia is not what it used to be with too many qualified scholars relegated to the ranks of proletarianized adjuncthood. Public service is hampered by a culture that devalues it and by occasional know-nothing anti-scientific administrations.
Debbie Kennett ,
>”On the other hand there are thousands of gun deaths each year in the United States. Why does the FDA not regulate guns?”
Questions like this indicate that you are clueless about US politics. Many Americans belive in the myth of Second Amendment rights. It goes back to something that some guys in breeches and tri-cornered hats wrote in the late 18th century and enshrined it in the us constitution. It is actually about a well ordered militia but five republican guys in black robes on the Supreme Court disagree with me.
Besides, it is the demonized elsewhere Treasury Department’s Bureau of Alcohol, Tobacco, Firearms and Explosives and not the demonized here FDA that would deal with your concerns.
Dear mrmr
thanks for the wsj report which I had not seen.
I do understand quite well how FDA regulation of IVDs works as it has been at the heart of my research for the last six years (see: http://www.kcl.ac.uk/schools/sspp/interdisciplinary/cbas/staff/acad/sh.html – I would be happy to share our publications with you should you wish to read any of them).
PMA is a hard road but then lots of companies travel that road and come out the other side. If DTC genetics companies are under-capitalised to meet the legal requirements for IVDs in the US, standards that other molecular Dx companies regularly meet, then that is unforunate for them, but I don’t think it is a great loss to anyone else. I do not see why DTC genetics should get a free pass and I do believe that DTC testing in general requires a higher bar (not necessarily PMA) as the assymetries of information between provider and user are generally far higher than for tests sold to highly trained professionals.
Whatever happens in the immediate future, it is clear that FDA regulation of IVDs is going to have to change if they are going to take on LDTs. DTC genetics is a very small (but rather noisy and highly visible) part of this much bigger picture and the FDA meeting on the regulation of LDTs was a much more important event than the Congressional hearing. FDA has, for a long time, said that it will have to be imaginative and flexible if it takes on LDTs and the regulated part of the industry is betting that as FDA broadens the scope of its regulation to include LDTs it will have to dilute the strength of its oversight – the agency just don’t have enough staff to put everything through its existing regulatory system to the same high standard. The proposed genetic test registry is likely to play a part in this emerging new regulatory landscape, and a new risk classification system may be part of the solution (one advocated by the industry body AdvaMed). Other solutions include a greater emphasis on postmarket controls and use of third-parties to carry some regulatory tasks (with FDA as ultimate authority or ‘meta-regulator’)
Finally: a lot of the hostility to regulation evident in many of the posts in this thread, and in the debate more generally, is not shared by people in the DTC genetics sector who accept that 1) regulation is inevitable 2) clearly defined standards will help industry in general 3) probably hope it will act as a barrier to market entry to companies who cannot meet those standards. Here is a quote from one of the founders of a well-known company taken from an interview I conducted in 2008:
“… this is an area that needs regulation and you know regulation probably isn’t under the auspices of any existing group you know, obviously in our country we have to change regulatory policy to get the FDA involved but clearly they should be involved, they should be involved both in the standard setting and both in the quality matrix and define what needs to be done because this is you know a healthcare decision …
The bottom line is that I don’t really care if it slows down our speed, I do think you need that hurdle. I think you’ve got a lot of players in the field that are high quality and you know if you’re high quality you shouldn’t be afraid of regulation and you know I don’t want people to be misguided by information so I’d rather be slowed by a year at [COMPANY NAME WITHELD] to get things done right in this country than not, even if they you know hurt us, it’s worth it. I don’t want the field to go away and that’s what’s going to happen if you don’t have the quality matrix.”
I see that the more hyperbolic arguments are still being made, even after a weekend to reflect on things. A little less “the sky is falling,” a little more actually thinking about the things that Stuart and others are saying, please. Just shouting that “the FDA won’t let me have access to my DNA” is nonsense. Shuren even stated in the hearings that FDA has no interest at all in non-medical claims. But the DTC companies are trying to have their cake and eat it, buy hewing to the “this is only for education or information” all the while offering medical advice.
Finally, and this might just be the cynic in me, name me an industry in the US that’s regulated, where regulatory capture hasn’t happened. Isn’t it bad enough that the past three decades have seen massive erosion of consumer protections?
Theres a common pattern here. The pro-government activists talk in abstractions about what regulation could possibly be. The ones in favor of genetic freedom talk about what regulation actually is.
Daniel how does the saying go? When you think FDA think TSA not CSI.
>”When you think FDA think TSA not CSI”
Hahh! TSA!
See:
http://www.cnn.com/2010/POLITICS/06/29/tsa.director/index.html
When you think TSA think Sen. Jim DeMint of South Carolina.
Email Gutierrez and Shuren to make your voice heard
dave, look, perhaps we can meet halfway on this. We both agree that the FDA’s work has been “less than stellar” and has “arbitrary timing”. And whatever my many failings, we both agree that I am but a simple person in a comments section, while the FDA and the GAO have a microphone that is heard by the whole world.
If nothing else, it is important that the officials in charge realize that their actions are controversial and that even their supporters believe they could have done things in a better way. Wouldn’t you agree that far-reaching regulation should have, if not consensus, then at least much more consensus than exhibited thus far?
In case you or others want to contact them, the relevant officials (in rank order) are Kathleen Sebelius ([email protected]), Margaret Hamburg ([email protected]), Jeffrey Shuren ([email protected]) and Alberto Gutierrez ([email protected]).
FDA is a law enforcement agency with police powers and should comport itself as such
dave, please consider another way to frame what you just said.
A government agency with a multibillion dollar budget has little or no idea how it should proceed. It has zero systematic scientific evidence showing that DTC (or LDTs for that matter) harms public health. But it decides nevertheless to “start swinging”, attacking small businesses (in the middle of a recession!), literally destroying jobs by scaring away investment and thus forcing layoffs. This attack triggers a Congressional investigation, which forces the startups to stop what they are doing to prepare reams of documents for the government. And then the startups start being trampolined back and forth between Congress and the FDA.
Unfortunately, it’s unlikely there will be a “bidding war”. These poor startups have no leverage to bid. Recall that the Senators and Congressmen at the hearing were already talking about what happens to their data after the startups go belly up, which is a foregone conclusion. Even Waxman acknowledged that PMA costs tens of millions of dollars and will likely put them out of business.
I ask you dave, in all honesty, why is it ok for the FDA to “grab a couple of big mallets and start swinging” with “little or no idea exactly how they will proceed”? Please remember that the FDA is a law enforcement agency with police powers. I don’t think anyone wants to live in a country where “start swinging” is standard operating procedure for the police.
But what standards? Remember,23andMe spoke personally with the FDA Commissioner in 2007, receiving his verbal ok before launching their product. I’m not sure if you recognize this, but there’s not a clinical lab in the country that knows what the standards for LDTs are anymore.
Try finding out the written changes in FDA policy on DTC tests from 2007 to 2010. The 23andMe launched in 2007 that the FDA Commissioner in 2007 (Eschenbach) approved is very similar to what they have today. From the very beginning 23andMe had disease risks, carrier status, ancestry, and drug responses. That was the best of the available research, but explicitly marked the whole time as “not diagnostic grade”. Just as a practical matter it’s not easy to validate all those variants on large samples.
Moreover, think about what you’re suggesting: why exactly should 23andMe have to come back to the FDA every time a new association is published to ask if they can incorporate it? You may not realize this, but you’re putting 23andMe into something of a catch-22. As Misha Angrist has observed, 23andMe is “luxury waste” if it just reports ancestry and hair curl, “unregulated diagnostic” if it reports Mendelian disease associated loci, and “not ready for primetime” if it reports on complex diseases. They can’t win.
One other thing to realize: for all kinds of structural reasons, most of what we know about genotype/phenotype associations was funded by NIH and is thus focused on relatively grim things like cancer and pharmacogenomics. Thus, in order to do the fun & nonmusical things, you need to allow something like 23andMe to exist, because NIH certainly won’t fund it.
But “figuring that out” means first “figuring out what is legal and how much it will cost to deal with the FDA bureaucracy”. No one knows what is and what is not legal. And no one knows whether that will change again in 6 months or a year or at the drop of a hat when a journalist writes another hit piece.
As for whether “complete raw sequence data is going to be available to everyone before long”, please remember that patients still don’t have access to their own medical records. So if the genome is considered solely medical, and if sequencing and interpretation is mediated by doctors only, then there will be no universal access to the raw sequence anymore than there is access to the medical record. The regulatory and legislative framework greatly affects what people think is “standard” and “reasonable”.
FDA did in fact declare DTC illegal overnight: a chronology
Perhaps we have different memories. As you may recall, Robert Stein of the Washington Post called OIVD’s Alberto Gutierrez and quoted him on May 11, 2010:
http://www.genomicslawreport.com/index.php/2010/05/11/pathway-walgreens-and-dtc-regulation/
As Dan Vorhaus noted, this was the first indication by the FDA that DTC testing was “illegal”. It was indeed declared illegal overnight. No law was cited, no notice of proposed rulemaking was put out, no Guidance was issued, and there was no quiet FDA contact with Pathway Genomics prior to giving a quote to a newspaper with international circulation. Gutierrez then went on to say in another interview:
So, after years of spit parties and New York Times articles, out of the blue the FDA decides to react on the basis of press coverage (“the fact is…[their] bold move…brought them to our attention”). And not really even “press coverage” per se, but Rob Stein’s specific question to the FDA, which he used to force them to abandon their previous “no comment” stance.
Perhaps the most alarming Gutierrez quote of all is this one from July 2009:
http://www.medicaldevicestoday.com/2009/07/under-new-leadership-fda-diagnostics-office-to-tighten-its-control-.html
So: in other words, there was no law passed. No public debate. Not even a changing of personnel at the agency. Just a top down directive from the administration to ramp up diagnostics regulation. And not just for startups, but for all clinical labs, including nonprofits, hospitals, and academic medical centers:
http://www.genomeweb.com/dxpgx/acla-meets-fda-head-propose-inter-agency-lab-regulation-laboratory-developed-tes
One really doesn’t see the logic here. Most of the lab techs I know are Democrats. So are most VCs and most highly educated people for that matter. They elected Obama to stop the war, not to go after clinical labs. God only knows why he’s beating up his base of academics, scientists, and nonprofits. All I can say is that Tea Party stickers are going up in the most unlikely places these days.
Ninety-ninth!
That’s just puerile, Jeff. Scoring the 100th comment, on the other hand…
[You’re both losers – ed]
At 05:08 AM 7/29/2010, Steven Yarrow wrote:
“So: in other words, there was no law passed. No public debate. Not even a changing of personnel at the agency. Just a top down directive from the administration to ramp up diagnostics regulation. And not just for startups, but for all clinical labs, including nonprofits, hospitals, and academic medical centers”.
No public debate?
In November 2008 the US conducted a presidential election and the Republican administration’s lax regulatory record was one of the issues. As Republican Sen. Lindsay Graham said recently on another occasion: “Elections have consequences.” Considering the necessary work to reverse decades of market-fundumentalist anti-regulatory erosion of public protections, our concerns here are a small part of the wider debate.
Would the general public (if it cares) favor increased diagnostics regulation?
I assume that industry lobbyists and PR agents are already working to influence policy outcomes.
Who is working on behalf of the citizenry at large?
To which stakeholders is the US executive branch responsive in this controversy?
Mr. Yarrow,
again, your hyperbole does you an injustice.
Nowhere did I say I support the FDA approach to this whole situation. I was just explaining what I saw, which is that the FDA needed a way to get the ball rolling in a serious way. And the ball is rolling. Was there another way? Probably. Maybe FDA pushed the ball off a cliff or in the wrong direction, but that remains to be seen. It is still very early in the process, and all your whining about foregone conclusions is premature.
Navi and 23 and probably others have been involved in closed door discussions with the FDA and NIH well before last week, and those discussions will continue. Risk based regulation of some sort appears to be on the horizon. The risks that a DTC test might impart, and the consequences if that risk is realized need to be defined. Where it is unclear that a defined risk truly exists, or what the magnitude or consequences of the risk are, it may be necessary to provide some provisional classification and ensure that ongoing studies to establish the level of risk are carried out. If companies really want to do research and contribute to science they will put their money where their mouth is and follow through. Better yet, there may in fact be NIH funding available for such work.
Whether you approve of the FDA approach or not, the “poor wittle industwy” (give me a break)is in need of some regulation. Even the industry acknowledges this. But even if the companies disagree with regulation, I don’t believe that effective rulemaking necessarily needs to meet with 100% consensus from those being regulated. The US govt has sold off many of the regulatory agencies to the industries in question, and I don’t personally (BP) find that to be particularly effective (Enron).
with rspect to DTC genetic testing,
Claims should not be misleading.
Potential harms and the risks of such harms when known should be disclosed. The risk data is largely still in the research phase (perhaps not with navi’s tests) – so treat risk disclosure the way we do in a research protocol.
Limitations of the risk estimates (not diagnostic, not precise in most cases, may exclude disease risks due to other known or known factors, highly subject to change, may be based on studies of ethnic groups other than yours) need to be made absolutely clear . Preferably pre-sale, if I was a customer.
I am not defending the FDA approach. But, long run, I think thoughtful regulation will help the industry and not hurt it. Last week’s congressional hearing did not demonstrate much in the way of thoughtful regulation, but the FDA meeting was much more encouraging, and the whole thing was the beginning and not the end of the story.
@Steve Yarrow
“. From the very beginning 23andMe had disease risks, carrier status, ancestry, and drug responses. That was the best of the available research, but explicitly marked the whole time as “not diagnostic grade”. ”
Steve, you are dead wrong.
A. They did not have drug response testing until last year
B. They did not have BRCA testing until Feb of 2009
C. They did not have the majority of carrier testing either.
I am beginning to see your worry, just not your embellishment of the facts. The facts are this
1. 23andMe are testing for medical grade conditions NOW
2. 23andMe et.al. are using CLIA labs NOW
3. DTCG are using clinically validated genetic markers for medical grade disease NOW.
4. The PGx chip approved by FDA is a medical device not an LDT
Dude, the writing is on the wall.
“poor wittle industwy”
This kind of tone is a real turnoff. 23andMe is a rare woman owned startup doing something innocent and fun. These women are not Halliburton.
@Jane
“23andMe is a rare woman owned startup doing something innocent and fun”
BRCA results are fun Jane? Really?
PGx plavix results are innocent? Really?
The only one innocent here is you. And naive to boot.
Woman Power! Yeah! Sexist……
When one of those founders talks about doping up her kid with Benadry, I hardly think she is innocent…. http://thegenesherpa.blogspot.com/2008/10/dtc-says-your-baby-genomeit-fun_3040.html
23andMe reported that I, a White woman of European descent, had… 0% European. The health report seems randomly generated. The traits report gives bizarre results. There are many haplogroups. After only taking samples from four population groups, they assign you to one. When I asked on the forums what happened, a bunch of bullies seemed to try to drive me to suicide. I want my money back.
AHJ is looking for partner websites in the medical genre. AmericanHealthJournal is a health site with over 3000 of high quality health videos. We are seeking blog owners who are interested in featuring our videos. We can offer content exchanges, link exchanges, and exposure to your brand. Contact us at our contact page on our site.
Howdy! This is my first comment here so I just wanted
to give a quick shout out and say I genuinely enjoy reading
your posts. Can you recommend any other blogs/websites/forums that go over the
same topics? Thanks for your time!