Posts Tagged ‘healthcare innovation’

Was the FDA Right to Shut Down 23andMe’s Marketing?

Monday, March 3rd, 2014

My opinion: the FDA was right to act in November by asking 23andMe to stop marketing  its personal genome service.  Now, don’t get me wrong, I have been rooting for 23andMe to succeed at what they are doing in the Consumer Genomics (CG) space.  I have been hoping that “opening the door to the genome” would produce some interesting new insights about biology that might escape traditional scientific inquiry.  I have been hoping, too, that taking genetic information directly to the people might further empower individuals in the health care system and begin to counter balance the medical profession and insurance companies.  However, once the FDA communicated to 23andMe that their offerings looked like a medical device to the FDA and would the company please take certain steps, 23andMe’s choice was really to comply or break the law.  Based on what’s public, the action by the FDA in November should have been no surprise to the company.

So, the real issue that many are unhappy about is whether CG should be regulated at all.  Should the FDA have categorized 23andMe’s personal genome service as a medical device in the first place? There is room for debate on that point, however, I think at that time it was a reasonable position for the FDA to take (remember, in 2010 the CG industry was not without controversy over quality and utility, at least according to the GAO).  Furthermore, the industry, including 23andMe was distributing what had for many years been considered medical advice.  Why should the CG companies get to ignore regulations that other diagnostic companies are bound by?  I, for one, am generally pretty happy to have an agency try to make sure that the medical products that I use are of reasonable quality.  That is not the case in other parts of the world.

Since that time things have changed.  Some studies have shown that CG info is probably mostly harmless.  It may well be time to engage more deeply in the conversation over what, if any, regulations should bear on genetic information.  While many are of the opinion that there is no harm to one knowing one’s own genetic code, I’m not so sure.  I can certainly envision in a large population some subset of people deciding to take their health in their own hands and making a bad choice based on erroneous genetic information.  Maybe we should let them do that.  Maybe that is really what we are debating in the guise of the 23andMe/FDA issue.

FDA Tells 23andMe to Stop Marketing: Death Knell for DTC Genomics 1.0?

Tuesday, November 26th, 2013

The FDA on Friday published a letter  addressed to 23andMe’s CEO, Anne Wojcicki, telling the company to cease marketing it’s Saliva Collection Kit and Personal Genome Service (PGS). In the letter, Alberto Gutierrez, Director of the Office of In vitro Diagnostics and Radiological Health, describes the many interactions the Agency has had with 23andMe, informing the Company of the need to comply with the regulations and attempting to assist the Company in doing so. He also describes how 23andMe did not fulfill their promises to provide information to the Agency. I’m kind of surprised that 23andMe did not work with the FDA, since it ought to be no surprise that the FDA would order 23andMe to stop, given their apparent lack of compliance. Hopefully we will find out more about why the Company thought they could ignore the FDA.

Two additional thoughts that this development seems to support: the final bell for DTC genomics 1.0 and FDA is going to regulate genetic tests. 23andMe was really the last prominent player in DTC genomics left standing, after Navigenics and deCode were purchased by big Life Sciences firms with no apparent interest in consumer genomics. It sort of confirms that the space of unregulated genetic testing with medical information packaged alongside is rapidly diminishing in size and probably won’t exist in this form for long. It seems likely that these technologies and products will end up being regulated and available largely via prescription. Score one for the medical profession. However, there may well be a next chapter written and it could have some interesting twists. Say, for example, a 23andMe equivalent offers that service from an international location. I think it will ultimately be hard to hold a lid on consumer genetic testing.

Where are we with Targeted Cancer Therapeutics?

Tuesday, September 17th, 2013

In a recent project, I received a draft manuscript discussing targeted cancer therapeutics and companion diagnostics in which the writer made this statement: ”CML patients in treatment with Gleevec now have the same life expectancy as people without the disease, while suffering few side effects”. I was aware that many CML patients respond well to imatinib (Gleevec), but wasn’t aware that imatinib effectively cured these patients, as this statement implies. If true, this is a stunning success. I am more accustomed to reports such as this paper on gelfitinib resistance or this one on imatinib resistance, which bolster the view that resistance to targeted therapeutics is essentially inevitable.

I took a closer look at the paper that this statement was apparently based on to convince myself. The publication that seems to have been the focu s of the attention that generated statements such as that above was published in the Journal of the National Cancer Institute in 2011. The first author, Caro Gambacorti-Passerini, and his colleagues analyzed outcomes in 832 CML patients who had been treated with imatinib. The figure that encapsulates this phenomenon is Figure 2 from Gambacorti-Passerini et al. J Natl Cancer Inst, 2011, 103:553, reproduced here:

Gambacorti-Passaro Figure

Panel A is the incidence for several types of leukemia/lymphoma, CML is indicated by black triangles; panel B shows mortality for the same diseases, with CML again indicated by black triangles.

Notice that the mortality of CML drop dramatically right around when imatinib was approved. We would like to do this with all cancers.

The caveat to the story that the writer in my project glossed over was that one of the enrollment criteria in the aforementioned study was that the patient be in complete remission at 2 years of treatment. That is to say, patients who did not respond to imatinib or relapsed prior to their 2 year anniversary were not included in the study. This criterion for participation means that this study examined survival of the patients with the very best response to imatinib. Now, it is still an impressive fact and excellent news for CML patients. It means that if you have a good response to imatinib, you are just as likely to succumb to another disease as to CML or, in other words, you have a normal life expectancy.

This got me to thinking. How are we doing with targeted therapeutics for cancer? Do other targeted therapeutics have a similar highly effective profile? Is the original statement that caught my eye just a narrow, subjective view of the situation for CML patients?

In upcoming posts I am going to take a closer look to see what lessons we might learn from this encouraging view of cancer therapy. I don’t think the effect will be quite as drastic for other targeted therapeutics, but I think it’s worth taking a look. Are those who respond well to some of these therapeutics essentially cured of their cancer? If not, what does the response look like and what does it say about where to go next. Stay tuned for future posts examining the impact of targeted therapies

Gene Patents No More

Friday, July 12th, 2013

As most of you know at this point, on Jun 13, 2013 the Supreme Court of the United States ruled essentially that native DNA sequences are not patentable subject matter.  The question ended up with the Supreme Court precisely because there are good arguments on both sides and, as you would expect, there was a lot of highly charged rhetoric exchanged leading up to the decision.  I’m not going to settle those questions here, but I did think it was worth a few moments musing about less technical aspects of patent issues.

I have tended to side with those who believe that gene sequences, at least the ones that exist in the body, should not be patented because they are a principle of nature—it just feels like giving too much away to me.  However, as one of the founders and early investors in Myriad Genetics shared in this article, without the monopoly guaranteed by a patent, investors would not have anted up to launch a company to develop the genetic tests, certainly not at that high risk time (circa 1991) when it was all but certain if these tests would be worth anything.  Others (e.g. this article have argued that the monopoly on the gene sequences that Myriad (and others, with respect thousands of other genes) have enjoyed have impeded progress in understanding genetic function and utility.

As with most persistent debates, both points of view are probably true in part.  From the perspective of the pro-patent camp, a large shift in how diagnostic products are developed that was necessary to bring tests such as these to market.  A huge promise of the human genome project was to provide for health care based on the sequence of an individual’s genome—personalized medicine.  But, it was a promise made, lo these many long years ago, when we really had no idea if it would actually work.  To get to market, we would need to generate a large amount of experience with these sophisticated new tests, the operating characteristics of which were largely unknown.  Would genetic prediction of cancer susceptibility actually work?  Would patients and doctors actually find the test useful?  Much different than measuring the number of white blood cells in a blood sample.

To get physicians and patients to order the tests, data on the validity of the tests was needed.  That required a lot of free testing to generate the data on the relationship between the gene variations and cancer incidence.  Myriad Genetics, as well as Genomic Health and other vendors of gene-based tests, have invested heavily in clinical validation of their tests in order to convince patients and physicians of their value.  Investors paid for much of this and patent monopolies were their reward.

On the anti-patent side of things, I wonder if Myriad Genetics has shot itself in the foot by jealously guarding its monopoly and appearing to be greedy (even after the Supreme Court decision, Myriad continues to aggressively pursue its perceived monopoly, see this article).  Stanford University and UCSF famously structured their genetic engineering patents to allow broad-based licensing, winning high levels of praise for licensing savvy and social conscientiousness.  Their patents did not meet the raucous challenges faced by Myriad.  In contrast, articles such as the one mentioned above  and this article decry the excessive costs imposed by companies of not only diagnostics, but therapeutics, as well.

Now I haven’t sat down and poured over Myriad’s (or anyone else’s) financial statements to ascertain if they really need to charge $4000 per test to recoup their investment and earn a reasonable rate of return for their investors.  But, what is clear is that that $4000 number is very high compared to what the world is used to paying for diagnostic tests.  It probably would have helped Myriad if they had been more transparent about why they needed to charge that amount, given that their soon-to-be competitors are planning to charge under $1000 in some cases.

Since that our government grants patent monopolies for the betterment of our society in general, I wonder if it might be prudent for companies and other patent holders to consider public reaction to how they handle the right to charge what the market will bear.  If the public perception is that the patent owner’s behavior is not in the best interest of society, they may be sacrificing goodwill, which ultimately, in closely watched cases, such as this one, might tip the balance one way or the other.

DTC Genomics Update

Friday, October 5th, 2012

I have fallen a bit behind in posts about the DTC genomics field.  One change in my posts to note with respect to this subject, I am going to start using the term “personal genomics” to describe the field, since it seems to better capture what these companies are about.  Here are some developments from the last few weeks:

23andMe patent for Parkinson’s

23andMe received their first patent titled “Polymorphisms associated with Parkinson’s Disease” in May.  As discussed here  and here this news reveals a bit more about the company’s commercial aspirations and creates a rub with their espoused culture of “democratization of genetic information”.

Navigenics bought by Life Technologies

On July 16 Life Technologies announced that it had bought personal genomics provider, Navigenics for an undisclosed amount of money.  It appears that LTI purchased Navigenics for its CLIA lab and to gain an entrée to genetic testing.  No word from LTI on pursuing DTC genomics.  What does this say?  Near term, even Navigenics thought personal genomics is going to be slow moving and an uncertain success.

23andMe seeks FDA approval

In July, 23andMe announced that it would seek approval for seven of its tests from the US Food and Drug Administration.  The nature of the tests was not specified.  As you will recall, in 2010 the FDA tangled with DTC genomics/personal genomics companies over the accuracy of their tests and related medical information.  This looks like another hint at the strategies 23andMe will use to actually make money in the personal genomics business.

Where does the industry stand now?  It is a struggling field, fraught with challenges, including relevance, uncertain social acceptance, and potentially a new level of government regulation.  Is it a field on the brink of extinction or explosion?

Lessons learned from tumor heterogeneity

Tuesday, April 10th, 2012

My recent blog post, Tumor heterogeneity, revealed…, discussed the New England Journal of Medicine article by Gerlinger and colleagues describing the genetic heterogeneity found both within a patient’s individual tumor nodules and between spatially separate nodules.  There has been a substantial amount of discussion of this work and angst about how it might signal the end of personalized medicine even before it really got started.  I don’t believe that will be the case at all.  To the contrary, this paper made interesting contributions in three conceptual areas that may help pull the field forward.  These areas are the 1) relevance of prognostic gene expression profiles, 2) the nature of “driver” genetic mutations, and 3) the pathogenesis of cancer itself.  All of these areas are, in my opinion, very important to make headway in before personalized cancer medicine can become a truly effective tool in medicine.

Heterogeneity in gene expression profiles across the tumor specimen

The result that most seized on to proclaim the demise of personalized medicine was the finding that gene expression signature from spatially separated parts of a tumor nodule yielded different assessments of prognosis.  The implication is that a single biopsy specimen is inadequate to generate an accurate prediction of clinical course or response to treatment.  Most likely that is at least partially true.  However, the issue is with sampling, rather than the molecular biology.  We have known for decades that tumors have variable histology within their mass, with some regions reflecting poorer prognosis than others via their histologic grade.  Rather than reflecting a conceptual disconnect that dooms a new paradigm, it looks more like a technical problem to solve, which should be no surprise along this new path.

Convergent evolution

Both the Gerlinger paper, as well as others (e.g. Walter et al, NEJM), using NGS have now demonstrated that within a single patient the same gene can be found to be mutated multiple independent times, suggesting that this mutation creates a change in gene function that participates in the development of the cancer.  This had not been shown in humans before.  This finding will be useful for clinical diagnostics  and it may be game changing in basic research.  In clinical diagnostics identification of a multiply-mutated gene would give additional confidence that the damage it represents is causal and may help select targeted therapy.  In basic research, identification of such genes would represent novel evidence of the causality of specific genetic changes in the disease process.  This type of evidence is a smoking gun, a sign post saying “Needs to be mutated to reach this disease state”.  This type of evidence, which only deep sequencing can yield, is a new and useful application of NGS that was not previously available.


The picture that the Gerlinger paper, Walters paper, and others paints is one of clonal evolution of cancer.  This type of work paints this picture with clarity that has not been achievable before.  What is striking to me is that these results make it harder to ignore the concept that these molecular alterations, as important as they clearly are in the progression of cancer, may not be the cause of cancer.  They beg the question, “what initiated this evolutionary process?”.  Certainly, oncogenes, tumor suppressors, and the like are a part of cancer pathogenesis, carrying the developing disease along.  But it seems to me that there is still a “first cause” of some sort that we have not put our collective fingers on.  Genomic instability is certainly key, but then what is the genesis of the genomic instability?  What are the inputs that kick this process off?  Efforts to answer these questions will move us closer to effective treatments for cancer and other diseases that may share these pathogenic processes.

An Academic—Industry Partnership to Study DTC Genomics

Tuesday, March 6th, 2012

A GenomeWeb story today provides details of a study being performed by researchers at Brigham and Women’s Hospital and the University of Michigan on the motivations for getting and effects of DTC genomic testing. The study will look at the attitudes and motivations of 1000 people who order tests from 23 and Me or Pathway Genomics before testing. These results will be compared with the subjects’ attitudes toward their health and changes in their behaviors following testing to ascertain how people use genetic information.

Two interesting aspects to this study are the academic-industry collaboration and the window into social attitudes toward genetic information. The ability to complete this research will require the industry participants to cooperate, which in turn required give and take from both academia and industry to accommodate the needs of both parties during the planning process. These two groups are often at odds, so it is heartening to see a partnership that recognizes that both parties are motivated to make positive contributions to the greater health good. The road to this agreement is described in this paper by Lehmann and colleagues .

Clearly, the payoff for this study will be answers to questions around why people are interested in this type of testing anyway and what effect it has on their lives. A big concern from the health policy world has been that dispensing this type of information without expert interpretation might lead to a range of ill effects on the recipients. Happily, this so far has not been the case and for the most part I don’t believe ill effects will be observed in the future.

I am curious to see what the motivations for so-called “recreational” user are, if those can be identified in this study. Similarly, it will be interesting to infer from these results how seriously people take genetic information. My sense of it is that these results may reflect back in surprising ways on the usefulness of a variety of other genetic and genomic testing services, primarily involving the risks of disease. This study will hopefully provide a novel window into what we as a society really think of genomics.

DTC Genomic Testing—What’s it good for anyways?

Friday, December 30th, 2011

What is the fuss over DTC genomic/genetic testing all about anyways?  DNA is just a sequence of letters, isn’t it?  Lots of people are experiencing angst over the fact that these upstart companies would have the nerve to sequence part of people’s DNA for them.  I mean, it’s just a bunch of letters, isn’t it?

Seriously, I have to admit, I, as a molecular biologist, have experienced a degree of self-righteous indignation that these so called entrepreneurs would debase the field of genomics and medical genetics by offering to sequence anybody’s DNA for a price.  It seems beneath all of the effort and concern that has been invested in developing the field.  All of that hard-earned knowledge being sold off the shelf like a cheap tabloid.  That was the feeling, anyway, and I imagine some amount of that type of sentiment contributes to the resistance to the development of the DTC genomics field.

However, the reality is that those letters are attached to a lot of other information that may have health implications.  There are several serious genetic diseases (ironically, most discovered prior to the genomic era) whose sufferers (or carriers) traditionally receive genetic counseling to learn how to cope with the situation.

Beyond these known disease situations, the hype of the genomic age has led to lofty expectations for genomics.  Those letters are our shorthand for the substance (DNA) that gives us our individuality and which when altered is may give rise to disease, tell us who our relatives are, and potentially make us weller-than-well (if only we can change it a little bit).  We’ve bought pretty heavily into the idea that we are our DNA and therefore, revealing it is, in a sense, giving ourselves away.  There is an ever-growing body of genomic information that pins many hopes and dreams and futures to those four letters.  So, it’s not surprise that feelings run high when it comes to genomic information.

So, DNA/genes/genomics is loaded with expectation, but what’s DNA sequence information really good for when one takes a hard look at it?  How is it being used now?  We can start with a partial list of uses that have been found for DNA sequence information:

  • Disease risk assessment
  • Disease diagnosis
  • Preconception screening
  • Forensics
  • Genealogy
  • Recreation

The fuss that these upstart companies have created has revolved around health information for the most part.  That would be the first three items on the above list.  These companies are seeking to sell their customers their own DNA sequence information, along with an assortment of linked information regarding the health implications of the DNA sequence in question.  It’s the health information being sold along with the sequence information that has caused the kerfuffle with the FDA and the medical profession.  And, for some understandable reasons…

Long before we even knew what DNA was, enterprising companies and individuals were taking advantage of our sensitivity around health issues, selling remedies and other noxious (or inert) substances to solve health problems.  This profitable, but unethical, behavior was addressed through creation of the FDA, whose job it is to keep the nation’s healthcare resources safe.  So, here we have what might be called the modern day version of the snake oil salesmen (at least in the estimation of some): the DTC Genomics companies.  Not surprising, then, that the FDA might feel compelled to step in, as it appears they are likely do.  Similarly, many in the medical community have allowed as to how they would prefer that their patients not have access to their DNA information.  Also not surprising, since for known genetic diseases the medical profession has heretofore controlled this information  However, as it currently stands, the genomic profiles being sold by DTC genomic companies are pretty innocuous, so it doesn’t stand to reason to restrict the type of genomic information the DTC companies are selling.

My view is that we stand at a crossroads of sorts.  Down one road we regulate human DNA sequencing as a medical procedure, bequeathing control of the resulting information to specialists licensed to dispense that information in carefully predetermined ways.  This is a suitable model when the dispensing requires extensive training to avoid injury to the receiving party, as in the case of prescription drugs or cardiac catheters.  For genetics in the current information-rich environment and age of patient empowerment, I believe that there are a limited number of situations in which harm would come to a person who knew their own DNA sequence.  And, even those cases (e.g. Tay Sachs disease) it is questionable if the actual harm is sufficient to bar access except under carefully controlled conditions.

The other road might be one in which one can obtain the sequence of their genome, if they are so motivated and can afford it.  It is likely that reasonable quality services will be available to provide this information soon (currently there are concerns about quality with many of the providers; note to DTC genomics companies: you would do well to pay attention to the quality of your sequencing if you want to survive).  The latter three items on the list above would be supported by relatively simple, low hurdle access to sequencing services.  In fact, my guess is that FDA regulations or no, in the near future a motivated person will be able to get their genome sequenced.  Somewhere.

My concern  is what we might lose if we over-regulate DTC genomic testing.   The latter three items on the list have emerged in recent years.  What else might be added to the list in the future?  What uses for DNA sequence data are not on that list?

What is DTC genomic testing good for anyways?  I don’t think we know the answer to that question yet.  Should we follow the Silicon Valley paradigm, let go of the information, and see what millions of “users” out there do with it?  Should we “crowd source” genomics?  Maybe there is someone out there with a marketing degree, a penchant for spreadsheets, and the interest in genetics who can offer a creative solution for the problem of missing heritability of SNPs.  Maybe a user group will surprise us by producing a creative solution to one or another vexing biology or health problem that has stumped the collective brain power of us professionals?  We may not know what DTC genomics is good for unless we give it a chance.

Are most published research findings false?

Friday, October 7th, 2011

Many people are aware of the work of John Ioannidis regarding the analysis of research findings and the conclusions drawn from those analyses.  In particular, these concepts were described by him in a paper published in PLOS Medicine in 2005 is apparently the most downloaded article from that journal.

I’ve had this article on my mental favorites list for some time now.  I am finally putting a few words in print about it mostly to put a stake in the ground on this issue because I believe it is an important one in this era of high volume research reporting.  In short, I agree with the article’s main conclusions, although I might phrase it as “most published biomedical research conclusions are not true”.  This is not to say I think there is some conspiracy or that statistics are useless.  To the contrary:  statistics is an enormously useful field of applied mathematics.  I also think a great deal of very good research is being done in labs and clinics around the world by very dedicated and smart researchers.

My concern over the veracity of biomedical research and how these results are reported stems from the nature of statistical models and test versus how they are interpreted and reported.  Within that discussion is another around the unspoken assumptions underlying both our biological and statistical models.

Perhaps the stickiest issue for me is the use, or misuse, of p values in many published studies.  Without getting too long-winded about it, far too often the p value is used all by itself and given the status of a “stamp of approval”.  Using a p value in isolation (i.e. p=0.001 therefore I won!) is ignoring a lot of important information.  What type of test did you “win”?  What distribution of p values for this test did you assume?  Are your assumptions correct?  Did you keep testing data until you found the p value you were hoping for?

Fortunately, I think the wider scientific community is waking up to the deficiencies in the most commonly used statistical analysis scenarios.  This recent article from Genomeweb does a nice job describing the basic appropriate role for statistical analyses in biomedical research.  An important distinction pointed out in their article is that statistical significance and biological (or clinical) significance are two different things.  When we rely on statistics to identify important relationships within a vast ocean of information, it is all the more important to understand what these mathematical tools are telling us.

As the wise scientist once said, “Never assume anything other than a 4% mortgage.”  I mentioned assumptions above in the sense of statistical models; assumptions also come into play in experimental design.  My sense of it is that these assumptions are usually underappreciated or perhaps even ignored.  The danger, of course, is that incorrect assumptions, statistical or experimental, can invalidate the results and conclusions of any research.  Often these assumptions difficult to verify, which we might be able to cope with, if we knew what these assumptions were.  Unfortunately, they are not part of the standard scientific reporting paradigm.  This recent article in PLoS Computational Biology sheds some light on the issue of reporting experimental assumptions.  Again, by bringing the issue to light there is hope that we can begin to change our science reporting procedures to incorporate some discussion of assumptions.

I find it reassuring that these discussions about accurate analysis and reporting of scientific research are surfacing.  Opening up communication about these critical issues will greatly enhance our ability to navigate through the ocean of biomedical studies available to us.

Genomic Test for your Kid’s Sports Ability? Oh, Please!

Friday, May 27th, 2011

The US FDA just sent out more letters to genomic testing firms asking them to explain why their testing kits should not be regulated.  The companies in question (and their target market) were Lumigenix (disease predisposition), American International Biotechnology Services (AIBiotech, workout optimization for athletes and also disease predisposition), and Precision Quality DNA (PQDNA, disease predisposition and drug response).

Based on the blogosphere reaction, the testing of genomic influences on athletic performance drew the most attention.  I don’t know much about genetic influences on athletic performance, but I don’t think anyone else does either.  Hence the reaction to such a product—is there really any value there?  I already have to submit a copy of my son’s birth certificate to enter him in certain sports tournaments.  Am I also going to have to submit his genetic profile so he can join AYSO?

For all three it appears to me that the FDA was pretty reasonable in exercising its mandate to protect the public health by blocking unreasonable medical claims for products.  It’s unfortunate for those companies that are trying to do the right thing by backing up their genetic testing services with real data; they may well have to carry the burden of federal regulation soon.