Angelina Jolie, Myriad Genetics, & patents on genes

By Razib Khan | May 15, 2013 12:11 pm

Because of Angelina Jolie’s revelation, the Myriad Genetics case is in the news again. If you don’t know what I’m talking about, look it up. Because of the patent Myriad can charge thousands of dollars for a test which would otherwise be much cheaper (and putting it out of reach of many without health insurance). My question here is simple: if you are a geneticist do you think Myriad’s position has any validity? The reason I ask is that I know many geneticists, and I know many geneticists read me, and I follow many geneticists on Twitter, but I’ve never encountered one who would be willing to defend Myriad’s position as plausible and passing the smell test. If you are one of those geneticists please leave a comment, because I’m honestly curious.

I went to the talks about the Myriad case at ASHG, and I have to say it was all law, and no science. The science was confused and laughable. The panelists themselves rolled their eyes and expressed resignation as to the garbled ratiocinations of the judges who reviewed the case. There is a classic “two cultures” problem.

MORE ABOUT: Myriad Genetics

Comments (33)

  1. Dmitry Pruss

    Not sure if I qualify as a geneticist, but since I occasionally contribute to this blog, I hope that you may let my comment stand. Two cultures, right, but so it is a two-cultures lack of understanding between clinical providers and research geneticists.

    Study after a study found that price of clinical sequencing tests depends on the genomic target size, and does not depend on exclusivity of patents. For example, a study by Cook-Deegan et al. is generally very critical of the genetic industry and potential impact of patents, but they find that in the area of Lynch syndrome (where wide competition between labs exist), per-kb prices are absolutely on par with BRCA testing (obviously, regulatory, reporting, and QA issues drive these prices much higher than is customary for research sequencing). The table is long, and I will paste it below, after mentioning that complete lack of insurance is less of a problem (because there is company-provided charity care for the uninsured) than poor quality of insurance plans for the indigent (although most US insurers cover genetic testing, state-run Medicaid’s are notoriously resistant). Another often overlooked problem is turnaround time: patients need their results quickly to guide their urgent medical care needs, and only industrial-scale labs can return results in mere days.

    Lynch syndrome sequencing / rearrangment analysis prices, from Cook-Deegan et al. :

    Baylor: $1,150 per gene or $3,200 for the MLH1, MSH2 and MSH6genes36

    Boston University: $2,995 for all three genes (MLH1, MSH2 and MSH6)

    City of Hope: $1,771.20 for MLH1, $1,474.56 for MSH2, $1,400.40 forMSH6

    Harvard: $2,700 for all three genes (MLH1, MSH2 and MSH6)37

    Huntington Laboratory: $1,200 for two genes (MLH1 and MSH2) plus $600 for MSH6 ($1,800 for all three genes)

    Mayo Clinic: $2,000 for two genes (MLH1 and MSH2) and $ 1,100 forMSH6 ($3,100 for all three genes)

    University of Pennsylvania: $1,360 for MLH1, $740 for MSH2 and $740 for MSH6 ($2,840 for all three genes)

    Quest Diagnostics: $2,940.00 for full sequencing of both MLH1 andMSH2 and $1820.00 for MSH6 ($4,760 for all three genes)

    Among for-profit testing laboratories, Myriad charges $2,950 for its COLARIS® test which includes full-sequencing of the MLH1, MSH2 andMSH6 genes as well as testing for major rearrangements. Rearrangement testing complicates the picture further, as each laboratory has its own price:

    Baylor: Rearrangement testing for either MLH1 or MSH2 is $625, rearrangement testing for MSH6 is not available

    Boston University: Rearrangement testing is included in the cost of $2,995 for sequencing MLH1, MSH2, and MSH6

    City of Hope: Rearrangement testing and dosage analysis for 7 exons inMSH2 is $547.56, rearrangement testing and dosage analysis for all exons in MSH 6 is $658.80

    Harvard: Rearrangement testing for MLH1 or MSH2 is $600, rearrangement testing for both is $80037

    Huntington Laboratory: Rearrangement and gene dosage analysis for both MLH1 and MSH2 is $600

    Mayo: Rearrangement testing is included in the above prices

    Quest Diagnostics: Rearrangement testing for both MLH1 and MSH2 is $540.00; Rearrangement testing for MSH6 is not available

    • Chad

      I would be very interested in knowing the profit margins of these tests. I could sequence these genes myself for ~ $100. That includes labor and ordering new primers that will never be used again.

      • Really? BRCA2 is 10 kb. Say 10 primers to sequence that (assuming you miraculously get 1 kb reads for each sequencing reaction), that’s got to be 50 bucks right? $5/primer is way cheaper than I get. Then about 100 for the sequencing reactions. Assuming all work the first time. Then you need to collect the sample, extract RNA, do an RT, then do your PCRs and clean up the products. Sounds like several hours work to me, maybe $20 an hour? $70 total, say. Then sequence analysis, you’ve got to look for a lot of different mutations, probably somewhat automated but surely it’d take at least an hour, that’s pretty skilled labor so call it 40. 260, throw in 40 for reagents and mistakes and whatnot, that’s 300. Not sure how much overhead to include, but I think most grants get like 50% extra for overhead, so let’s say $150. $450 total. Say $250 total for BRCA1, it’s much shorter. $700 combined, and I was trying to be pretty optimistic. Of course the primers will be negligible if you’re doing lots of these, so let’s subtract them and call it $600. Not $3000, but not cheap either. Maybe cheaper if you automated a lot of steps.

        • Chad

          Point conceded. I admit ignorance on the rather extreme lengths of the BRCA genes, particularly when the average human gene length is on the order of ~3kb, more typical of the genes I usually work with. At these lengths costs are certainly cheaper. However, your costs do seem a bit on the high side to me. Ten sequencing rxns would cost me $60 while primer costs are on the order of $6-7/primer. Perhaps this is a reflection in differences of institutions. I also am not fond of “kit biology” which really does not save much time, but certainly slashes the costs in reagents. Not to mention most of the time here would actually be spent waiting around for the RT or PCR to finish, time that could be better spent doing other work.

          • You could probably cut down the prep time a bit, after all you’d want to do them in parallel, not one a time. And probably other places as well. OTOH, this would just be a quick and dirty sequence. I have no idea what the standards are for clinical sequencing, except that I’m pretty sure they’re significantly higher. Surely they’d want to get each sequence done two or three times at least. I dunno. Your point probably still stands about profit margins, even if it cost $1000 they’d be doing pretty well for themselves.

          • Chad

            Well consider my original estimate foolish based on the standards of research versus clinical.

      • Dmitry Pruss

        Net operating margin, 24% Q1. You grossly underestimate the size of
        these genes. Thousands of distinct mutations, dozens exons, about 20 kb
        of cDNA – for $100 in rxns and labor? Give me a break. + for clin apps,
        the same reagents are priced at a premium, and the personnel has to have
        very high formal qualifications. The equipment also comes at a premium
        to be certified for the clinic, so you take a hit on depreciation, too.
        Then insurance won’t cover your test because a more sensitive test is
        avl. (who covers sigmoidoscopy anymore when more expensive, but more
        sensitive colonoscopy is avl.?). So it will cost patients more, out of
        pocket, than Myriads. Then you bill them and they won’t pay. Then you
        make a mistake costing someone a cancer missed, or a breast needlessly
        removed, and they sue. Out of business.

        That’s why there are no discount providers in Cook-Deegan’s report:
        far too expensive to run a clinical lab even in the days of cheap
        primers, because the primers are such a minor part of the bottom line.
        There is a lot more than the lab costs: regulatory compliance,
        development and validation costs; liability and legal; billing and bad
        debts; marketing and physician education; etc.

        Of course in the
        US health care industry, there is no market play of supply and demand.
        The lab test prices are defined by procedure codes, out of understanding
        of clinical costs. Patents do not make things markedly more expensive,
        but patents do increase marked share and test volume, and therefore
        margins of patented tests may be higher even though the prices are

  2. svman

    If I have a gene data file, from 23andme or another vendor, can I easily find out whether I have any of the faulty BRCA genes without paying Myriad? How?

    • razibkhan

      23andMe has only a few SNPs to identify variants. last i heard their analysis mostly benefits ashkenazi jews. but there are other groups with the variant (jolie is not jewish).

      • Melissa

        Could you use imputation?

        • Melissa … I believe the short answer is no on imputation using raw 23andme data. You fail to provide any information about cancer in your family.

          Even if you “could do” something with imputing 23andme raw data, you would still fail as you would be lacking access to the Myriad database … and that is more of a competitive advantage for them versus their patent.

          Here are some comments yesterday from 2 posters on the 23andme site:

          QUESTION (Posted by HelenKathleen in Health):

          SUBJECT: “Usefulness of BRCA tests for non-AJ customers”

          I do not have the 3 BRCA mutations or Chek2 deletion tested by 23andMe, but my mother died aged 60 of ovarian cancer, and her twin sister died a few years later of breast cancer. Does anyone know how useful the 23andMe tests are for someone like me who is not Ashkenazi Jewish?

          ANSWER (mewahl):

          Do not – I repeat, do not – use 23andme’s BRCA testing as a replacement for Myriad’s service! Yes, there is a special place in hell for companies whose gene patents keep life-saving diagnostics out of the hands of the poor, but there is truly no substitute for their BRCA test, as I’ll explain below.

          There are many different alleles of BRCA1/2 that render the genes non-functional. Some of them are SNPs, and therefore testable by 23andme. However, many of the mutations are deletions of varying length and position (and when you think about it, isn’t this a much more obvious way to render a gene non-functional than mutating a single basepair?). SNP arrays like the one 23andme provides DO NOT detect for these large deletions. Therefore I would not put any stock by 23andme’s failure to detect BRCA mutations in your genome, Ashkenazi or not.

          The test that Myriad provides, which is actually dozens of procedures packaged together, would check your BRCA genes against a huge DATABASE OF KNOWN MUTATIONS that has been several decades in the making. I don’t agree with Myriad’s ethical decisions and certainly don’t have much money to spare, but if it were me in your position, I would trade in my car for a sack of cash and *walk* to the hospital for the BRCA test.

          • Melissa

            The health history of that side of the family is largely unknown. My grandfather who is the source of the Ashkenazi pretended to be non-Jewish and separated himself from his family. I am lucky to have fairly decent insurance so I will ask my doctor about the test.

          • Dmitry Pruss

            Melissa, even good insurers would typically balk if there isn’t much known about family history. Then if you claim Ashkenazi heritage then coverage will be easier to get, but only for the Ashkenazi BRCA mutations. You may want to talk to a certified Genetic Counselor in your area to figure out how to navigate these waters, and best of luck.

            When the family history is a blur … ahh. For people like myself, who have much cancer in the family but nothing like a well-known genetic cancer syndrome, and who have whole branches of the family decimated and lost in wars and upheavals, even today’s exome may be the best answer there is. And I got mine through 23andMe / BGI, but 1000 bucks and countless hours of searches and analysis later, no magic.

  3. Chad

    Its seems like this case has been going on for a decade or more. I find it very hard to take the justification of Myraid’s patent seriously. My only concern is the potential repercussions should it be struck down. I do not foresee this as a major impediment to research into new the genetics of disease or the identification of risk variants. But if this has any application to the patenting of products of genetic engineering, then this could have very negative impacts.

  4. I would argue that the most relevant experts would be medicinal chemists, not geneticists, because a much larger percentage of them have experience with patent law as it relates to chemical products. I’m inclined to just ignore anyone’s opinion on the Myriad case who isn’t familiar with the concept of natural product patents. The science is a lot simpler to understand than the law, in my view, but maybe that’s just because I find genetics more comprehensible than law.

    I am not a chemist, and I’m not sure if I count as a geneticist (though my lab is as much a genetics lab as it is anything else). I think Myriad has a case by the letter of the law, but it may violate the spirit of the law. The molecules they’ve patented are different from what occurs in nature- they are isolated, purified, and concentrated, and that’s what makes them useful as diagnostic tools.

    • Chad

      How broad/narrow is the definition of “isolated, purified, and concentrated”? Nearly all next-generation sequencing technologies requires one to first fractionate the DNA, attach adapters to the fragments, amplify by a few rounds of PCR, attach primers to the adapter sequence, and then finally sequence. The specific BRCA1/2 sequence would not be isolated or purified, but it would indirectly undergo quite a bit of manipulation if we now consider an unmodified PCR fragment sent for sequencing to be somehow unique from the natural state. In human next-gen sequencing, it is very popular to sequence the exome, in which case exonic regions are first enriched by hybridization to a chip and then sequenced. While BRCA1/2 would not be enriched in comparison to the other exonic regions, it would be enriched greatly compared to the entire genome.

      So at which point can we say that one is not isolating or purifying BRCA1/2? If I ran a very long PCR reaction or cloned a region of several kilobases that included BRCA1 or 2, along with other genes and then sequenced this whole region, would I be violating the patent?

      The idea that a PCR fragment alone is patentable seems absurd to me. Now if I were to modify that fragment, put it back in humans, or mice, or yeast or whatever, then that is something very different.

      • It’s a good question, and the ACLU claims that deep sequencing must inevitably infringe on gene patents. But it’s pretty easy to argue that, even with exome sequencing, you’re not targeting the BRCA genes specifically. Sure, they’re enriched relative to straight genomic DNA, but are they enriched in any *useful* way? My (non-lawyer) interpretation of the isolated products rule is that the isolated product must be substantially different in utility from the molecule as it exists in nature. A tube of DNA enriched for exomic sequences is not substantially more useful for identifying BRCA mutations than unenriched genomic DNA. So, for deep sequencing, I think the argument for non-infringment is strong enough to allow deep sequencing and gene patents to coexist.

        But your second point is harder to answer. I’ve looked at the BRCA patents a few times over the past year, and I’ve never been able to find anything that actually defines, quantitatively, what counts as an infringing molecule. Now, the BRCA genes both span something like 100 kb, IIRC, so you wouldn’t be able to PCR the whole thing, and I figure any PCR reaction amplifying a useful portion of the coding region would be obviously infringing, but if you’re cloning a neighboring region, and you happen to clip off a few kb of the BRCA region, would that infringe? If you clone a few hundred kb into a BAC or YAC, and it includes some or all of a BRCA gene, does that infringe? I dunno. I’m not sure what practical use that would be, I suspect it would be easier and maybe even cheaper to just sequence the whole damn genome, but it does seem unclear to me.

        As for whether PCR fragments are patentable: do you think human insulin should be patentable? It was patented, back in the day, without controversy. What makes one isolated human biomolecule patentable, and another not?

        • Chad

          I see a key difference with human insulin. In the case of insulin production, you have a cloned gene that is truly modified in the sense that it has been inserted into a vector for production in another organism.

          I see a distinct difference between patents which apply to an individuals ability to understand their own sequence as opposed to the modification of this sequence for commercial production.

          To understand where I draw the line. I’ll take something that I am more familiar with, the 5-enolpyruvyl shikimate-3-phosphate synthetase from Agrobacterium which confers resistance to Roundup in plants. If I were simply to go out and sequence this gene from wild strains of Agrobacterium, then I honestly do not know how we could logically consider that an infringement on a patent. Now if I were to try and mimic Monsanto’s technology and modify this gene, put it in other organisms, etc, that is obviously an infringement on ones patent.

          • Ah, but insulin was patented long before cloning. It was the purified protein that was first patented, way back in 1923. Only difference between the patented molecule and the molecule in nature was that it was purified.

            I, too, see a difference between patents that mainly apply to understanding an individual’s sequence and patents that apply to making something of direct use, but I disagree that that difference is distinct. To take your example: rather than the gene, let’s just talk about the enzyme 5-enolpyrwhateverase. Let’s pretend the purified enzyme is somehow useful. Should that be patentable? If not, what kind of purified natural molecules should be patentable, if any? If it should be patentable, why that, but not the purified DNA molecule encoding the enzyme?

          • Chad

            Certainly this is a grey area. I am no lawyer, however, I still see distinct differences.

            Does the patent to human insulin apply to the insulin my own body produces? Does the patent holder have the right to grind me up and extract that insulin (I am being hyperbolic)? Does the patent holder have a right to charge me for my body producing insulin? I think we can agree that these are necessarily ridiculous.

            My genome is more than my property, it is part of my personhood. Its one thing for patent rights to extend to a natural product extracted from a plant, bacteria, or animal. Its quite another thing to extend it to me as an individual, to you as an individual, or to Angelina Jolie and everyone else.

            What Myraid’s patent does, that neither Monsanto’s nor the patent holder of human insulin does, is deny us as individuals the right to our own bodies, our own genome.

          • Anthony_A

            The way I see it, a naturally-occurring material shouldn’t be patentable, but methods for synthesizing it, or for extraction and purification of it, should be. So human insulin, not patentable. Method to synthesize human insulin, patentable. Intermediate compounds in the synthesis, if not otherwise found in nature, patentable. Method for extracting and concentrating sheep insulin to use for producing human insulin, patentable. Sheep insulin, not patentable.

            Unfortunately, this is not the way that current patent law sees the issue, and there’s not a lot of hope for either legislation or court decisions which will institute what I think is the sensible way to deal with natural materials and patents.

          • Dmitry Pruss

            Before insulin, the first purified hormone to reach the Supreme Court was adrenaline, in 1911. Per Judge Learned Hand’s decision, it’s not about purification per se, it’s about enabling whole new practical uses which weren’t possible with the natural form of the compound. In case of adrenaline, the natural form was an adrenal gland extract; the purification made it safe for IV injections for the first time.

            Pure forms of BRCA1/2 DNA (such as synthesized primers and probes, PCR amplicons, cloned DNA etc.) enable such practical uses as sequencing, expression analysis, copy number analysis etc.

  5. svman

    I am not concerned about the extremely rare mutations. I wonder whether Jolie could have detected her problem based on SNPs from a cheap DNA test.

    • What I’m saying is that most of the mutations are quite rare. There are many many different harmful mutations known, and given that these mutations are fairly rare to start out with, each individual mutation must be even rarer. I don’t know how many of those known mutations are included in the 23andme test, but I suspect a lot of them aren’t. Especially since not all of them are SNPs.

  6. Dmitry Pruss

    Jolie probably got the mutation from her mother
    and she would have got tested just for the mutation found in her mother, too. Specific-site clinical testing is inexpensive (about $300), fast, super-sensitive, and widely covered by insurers. Since BRCA mutations run in families, after the first expensive positive result which looks all over the place, the rest of the relatives already know where to look.

    If you sequenced your exome, you can test every protein for frameshift and nonsense mutations, not just BRCA. This fails to cover some introns that Myriad sequences for no apparent reason
    Clinical exome is expensive, has gaps due to non-uniform coverage, misses introns as you already noted (where hundreds of disease-causing mutations are found – ever heard of splicing?), commonly miscalls frameshifts, fails to see genomic rearrangements, and most importantly, can’t make much sense of missense mutations (and some of the most common cancer mutations are missense). In addition, there are nonsense mutations in BRCA genes which do not cause cancer at all. You may need better coverage, better indel calling, introns and UTRs, rearrangements, and a good mutation effect database to get a quality test result.

    Exome is also painfully slow. If you have a breast tumor and a surgery scheduled next week, and you gotta decide between lumpectomy or mastectomy, then you needs an answer right now, not in months.

    Of course over time some “exome plus” approach might become “the way”. And for many mystery conditions, like unexplained child development abnormalities, it’s already becoming the standard. Sure, it’s slow, tedious, pricey, and misses many answers – but there are situations when there isn’t anything better. But for a well-studied condition like hereditary breast and ovarian cancer, target sequencing of specific genes will remain to be answer for quite some time.

    • Douglas Knight

      That’s a good point that the cost is $3000 per family, not per individual. The only quote I could get for single site brca is $600, though.

      I don’t think any other of your points are relevant.

  7. Neuroconservative

    I am not a geneticist per se, but my research involves genetic associations and I am quite familiar with the science & law of the Myriad case.

    Without getting into the details of the case law, which I think has taken a series of wrong turns over the years, I am generally supportive of Myriad’s position — not its specific legal position in the current litigation (which I think has painted them into an untenable corner), but its broad moral position as an inventor that has a right to protected opportunity for remuneration. Of course, I also believe that Mayo v Promotheus was decided wrongly:

    It seems to me that a lot of the animus on this issue comes from leftists (including most academics/geneticists) who are offended that someone would seek to make profits from other peoples’ health. This is amped even further in the public mind by the GATTACA-panic summoned by the notion of “patenting genes.” (Razib, I know that you are neither a leftist nor a genetic exceptionalist in the GATTACA-panic sense, but I am speaking generally).

    I am fully aware that, on policy issues, I am an outlier in academic science. It may also be true that new technologies are making any such patents virtually unenforceable. But, as a believer in the promise of personalized medicine, I am dismayed by the fact that the legal regime seems intent on telling venture capital to avoid this area like the plague.

    • Mayo vs Prometheus? Really? While I tend to be more on the pro-patent side, that one seemed obviously ridiculous. I will outsource my reasoning to one more versed in patents, and probably more sympathetic to patents than me, in general:

      “The Prometheus patents comprise these steps:

      1. Telling a doctor to administer the drug to a patient.
      2. Telling the doctor to measure the metabolite levels in their blood after dosing.
      3. Describing the upper and lower acceptable bounds for these metabolites, and telling the doctor that these indicate a need to raise or lower the dosage.

      There! That wasn’t so good now, was it? I railed against this ridiculous idea at the time, which is tantamount to trying to patent the entire practice of medicine, step by step.”

  8. Neuroconservative

    Addendum: A few years back, I had a preliminary finding that I thought had the possibility of leading to IP. I can assure you that I worked harder that year than I have ever worked in my life. Ended up with a couple of nice papers, but nothing patentable, alas. Nevertheless, I am a strong believer in the incentivizing role of IP protection; perhaps more importantly, my dean is too!

  9. Dmitry Pruss

    Throughout the decades (!) of Myriad patent challenges, the aggrieved parties stayed the same but their “reasons” to overturn the BRCA patents kept changing! First – “Others contributed so profoundly to the discovery that the patent is rightly theirs, not Myriad’s”; then – “it was a trivial observation rather than a discovery”; then – “DNA is a text and must be covered by the 1st Amendment rights”, and now “patent other molecules as you please but make a special exception for isolated DNA”.

    It is a legacy of an acrimonious conflict between the groups racing to discover BRCA1 and BRCA2 in the 1990s, and the losing side has never stopped fighting in courts. But after losing fight after fight, and changing the story line time after time, how can it look sincere now? Does the argument against the patent have that much validity, if its tenor has radically changed so many times?

    There may (or may not) be a dire public heath to un-patent BRCA1/2, but the US Government already can to do so, exercising its march-in rights under Bayh-Doyle Act. Basically the discovery was made by a public university and paid by the Fed grants; the University was required to patent and commercialize its discovery to recoup the taxpayers’ money; but the Fed reserves the right to march in and take over the patent if a public need requires it. So if the reason to un-patent is dictated by public health, then there is no need to go to Supreme Court; the solution is in hand already.

  10. “Could you patent the sun?”

  11. George Jones

    This was in the New York Times today about Myriad and has some relevant opinions and information.

    David Agus was one of the founders of Navigenics. He says:

    “Without competition, Myriad can effectively charge whatever it wants. Later this year, the company said it would begin incorporating the BRCA test into a 25-gene cancer-risk evaluation and phase out the à la carte BRCA test by 2015. For that broader test, it projects a gross profit margin of 87 percent.”

    “I’m all for innovation and the right to protect intellectual property, but when there is a clear monopoly and human lives are at stake, we need legislative action for rational and appropriate pricing. We don’t make vaccines prohibitively expensive so only the rich can protect themselves. Nor should we let other preventive measures that can save thousands of lives be priced at levels far above what normal “market conditions” would suggest.”

    Odds are good that the Supreme Court will side with Myriad … see recent case where there sided with Monsanto.

    It’s then up to Congress to set things right.

  12. Richard Seiter

    Any thoughts on the “Lander brief” amicus brief that Eric Lander filed? (contains a link to the actual brief)

    Regarding quantitative standards for infringement, the 15 consecutive matching base pairs standard seems stunningly (as in what was the PTO thinking in approving it) broad:

    There were quite a few amicus briefs filed for those who would like to dig deeper. For example:


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar