Pharmacogenomics Reporter has a fascinating article (subscription only, I think) on the impact of individual gene patents granted by the US patenting system on the future of personal genomics. Essentially, the issue for companies conducting genome-wide analysis (including SNP chips or whole-genome sequencing) is that setting up licensing deals for each individual gene makes business complex and expensive - potentially discouraging investment in the field.
On their website, personal genomics company Navigenics explains the problem:
For example, if we obtain licenses from third parties to 10 patents, each covering the use of one SNP included in our service, and each subject to a royalty of between 1 percent and 5 percent of our net sales of the service, we would be required to pay between 10 percent and 50 percent of our net sales revenue -- just for gene patent licenses! [my emphasis]
The PGx Reporter article lays out the strategies being employed by various companies to deal with the problem. For instance, Navigenics has come up with its own "universal royalty model" that divides up the amount it pays for licenses according to this formula:
This approach caps the total licensing fees payable by the company to 5% of its net sales revenue from the licensed service.
The article also interviews Jorge Conde, CEO of whole-genome sequencing provider Knome. Conde takes the opportunity to define the service his company provides in as unambiguous a fashion as possible:
"Patent pools [would] ... lower the transaction costs and simplify licensing especially for companies that are doing comprehensive analysis and not doing clinical diagnostics," [...] "If a personal genomics firm wants to include an IP-protected association just for screening processes, not for diagnostic or for clinical purposes, [there should] be a central marketplace or a clearing area..."[...]
"I think it would be beneficial to differentiate between companies that [use the] IP for clinical and diagnostic purposes and companies that want to use the associations for general screening purposes." [my emphasis]
There's no subtlety here: Conde wants to clearly differentiate personal genomics companies like Knome from "clinical" or "diagnostic" genetic testing.
That's a mighty fine (if not non-existent) line to tread, but it's one that all personal genomics companies have been walking very, very carefully ever since the field emerged barely a year ago. On the one hand, they want to sell their product as something relevant to consumers' health; but on the other, the looming hand of regulators means they are desperate to avoid branding themselves as selling medical diagnostic tests. That makes for a balancing act that is either entertaining or enraging, depending on your perspective.
The only company in the personal genomics space that seems happy with the patenting status quo is deCODEme - but that's hardly a surprise, given the number of gene patents owned by the company's mothership, deCODE Genetics. deCODE's central role in modern disease genetics is the major selling point deCODEme pushes to differentiate itself from its jazzier rival 23andMe, and the PGx Reporter article suggests that this may prove to be a successful strategy:
Over the past year, [deCODE] has launched clinical diagnostics for myocardial infarction, glaucoma, atrial fibrillation, type 2 diabetes, and breast and prostate cancer. The SNPs associated with these diseases, many of them discovered by Decode, are screened under the DecodeMe service.
However, the patents for these SNP associations have not yet issued in the US. When they do, the company expects to have an advantage over competing personalized genomics firms.
How much of an advantage that provides is yet to be seen - indeed, it seems to be an open question whether deCODE will even survive the winter, given its dwindling capital reserves and the precarious state of the Icelandic economy. The strategy might even backfire if it undermines deCODEme's ability to play the "not really medicine" card to regulators.
So, what is to be done about the gene patenting system? PGx Reporter notes that a US government committee has released a preliminary report on the topic, which considers a range of measures to fix the looming problems:
[The committee] is currently considering whether to propose a ban patenting gene/disease associations; restrict DNA sequence patents entirely or limit them based on specific use or for diagnostic purposes only; modify the Patent Act to restrict the ability of individuals to sue companies for patent infringement if they blocked access to a genetic diagnostic; and exempt from patent infringement litigation doctors and researchers using genetic tests.
The US government clearly needs to act soon to resolve the uncertainty swirling around the young field of personal genomics, but there's surprisingly little urgency about the committee's progress: despite working on the report since 2004, the committee will only be releasing its draft for public comment next February and it won't be until the end of 2009 that the final version is submitted. As the industry (and consumers) wait to see how regulations will change, technology marches on - there's little doubt that Knome won't be alone in the large-scale sequencing space for much longer.
So, with regulatory changes on the horizon, a financial crisis limiting consumer spending on luxury goods, emerging competition and an explosion in sequencing technology, the consumer genomics industry looks set for a tumultuous 2009.
- Log in to post comments
I wondered about this back in september:
http://scienceblogs.com/geneticfuture/2008/09/your_own_private_genome.p…
and I think my question about avoiding IP problems by just handing people their raw data is still valid:
DeCodeMe seems like an inherently bad deal: they can basically only test for SNPs that are currently associated with a disorder but aren't covered by someone else's intellectual property rights. If their risk assessment or choice of SNPs is crap, then so is your analysis.
Knome sounds much more promising (if they do actually sequence your whole genome), but it gets into some legal gray areas if they do.
If you sequence someone's BRCA1 and BRCA2 alleles as part of a whole genome sequencing, analyze the results, and then give them a risk assessment for cancer, Myriad Genetics will shut you down for violating their patent.
So I'm certain Knome leaves out a lot of the most valuable analysis, that or most of the $350,000 is going to royalties instead of sequencing.
But what if you sequence someone's genome and just hand them the raw data? If they subsequently use BLAST or another online DNA database tool to find out the results on their own, are you in the clear? Not many people know how to do that (or score the results), but an "independent" party could easily distribute a free application that acts as a front end to BLAST. People could just dump a sequence on the icon and out would pop out their BRCA1/BRCA2, APOE, and other risk profiles.
What about companies that perform sequencing using whatever samples and primers they are given and then just mail out the raw data? Are they responsible for making certain they're not inadvertently performing an expensive medical test?
I'm with Michael Crichton on this one (if nothing else). It is ridiculous to claim ownership over a fact of nature. It would be like patenting stars. No one is allowed to look at it without paying the discoverer royalties. If you want control over who looks at what stars, build a better telescope and patent it, otherwise GFY.
"No one is allowed to look at it without paying the discoverer royalties. If you want control over who looks at what stars, build a better telescope and patent it, otherwise GFY."
That is how the genome is patented now. The patents aren't for the sequence itself, but a "method and application using DNA" that specifies a process to determine something about the body using DNA, equipment, and specific oligonucleotides. I believe what happens is that there is a patient "template," and one simply swaps the oligonucleotides and other primers to effectively (but not literally) claim parts of the genome.
It's like patenting a specific way to point your telescope at the night sky rather than patenting the star itself.
re: formula. This will fall apart quickly when there are 3 snps that all affect disease X. SNP A accounts for 50% of the risk. SNP B accounts for 10% of the risk. SNP C accounts for 5% of the risk. The company that owns A is not going to be ok with making the same amount of cash as the owners of C.
For self serving reasons, I'm all in favor of everyone releasing raw data, but if that's all they do, there is no way for services to distinguish themselves, and snps will be a commodity with a race to the bottom for price. That is not a longterm business strategy.
The telescope analogy is a correct assessment of what patents are for, but I don't think it accurately captures they way patents have played out. I don't want to use myriad's telescope, I only want to see the star they discovered using their telescope. And I've got newer better ways of doing that without their telescope. To the patent office it seems that doesn't matter, Myriad still own the star for the purpose of 'looking'.
hibob: see http://www.promethease.com
hibob - I see Mike has already linked you to his application Promethease, which is the best available example of a third-party app to explore SNP data from personal genomics companies. However, I'm pretty sure anyone who provided an app to interpret BRCA1 sequencing data would fairly rapidly find themselves on the end of a Myriad lawsuit; so if the system doesn't change, third-party apps would have to be underground and anonymously distributed, which isn't necessarily a reliable way to be harvesting medically relevant information.
Andrew - are you sure? IANAL, but from what I've seen the patents seem to be interpreted as covering "any way of using genomic region X to provide predictive information about disease/trait Y".
Mike - I agree that weighting the licensing fees by the fraction of disease risk captured makes sense (although these figures are fairly shaky for a lot of common variants). Jeffrey Gulcher from deCODE makes a somewhat similar point in the PGx Reporter article, albeit using a rather self-serving example:
Matt - the "genes should be free" ideal sounds nice, but the side-effect is that there is a reduced incentive for investment in gene-finding efforts. Perhaps that doesn't matter - maybe we should just rely on government funding for gene-hunting, while corporations focus on developing diagnostic technology. If this is the decision society makes, so be it, so long as it is made with the understanding that this will have a negative impact on innovation in the field.
No, I'm not sure, and I'm not a lawyer, either. I think that the methods and application patented is effectively "any way," but not exactly, because "any way" is not a patentable statement.
Keep me updated if you find something more specific.
Also, you need a specific application. You couldn't "effectively patent" a star because there are no known applications. (too bad, astrologists) I would note that no patentable potential has not notably hindered astronomic discoveries.
I didn't say the weighting made sense. Doing so certainly simplifies the consumer's life. I just assume the owners of the most valuable IP won't pool it at volume prices, since its probably not in their best interest.
A long term consequence I can imagine is that certain tests are considered the gold standards, because they test for true causal variants. These will cost big money and be paid for prior to taking definitive action. Most folks will instead be tested for nearby proxy snps akin to the way that 23andMe sugests using rs3751812(T) in place of rs9939609(A).
Particularly bad news results will then be reconfirmed against the causal snp. This will create a wide market for low cost testing, with a healthy niche for owners of causal snps.
Gene patents are process patents since you can not patent a discovery or scientific fact. Thus, I believe that the current patents do not have a legal leg to stand on in the personal genomics era. For instance, when you buy a SNP chip you obtain a license for using the chip, primers, and associated intellectual property utilized to detect SNPs. You can then genotype away. DeCODE is not Illumina. They can not prevent you from genotyping one of "their" SNPs nor can they demand that Illumina pay a license fee for selling a chip with one of "their" SNPs on it as long as Illumina does not "read" the SNP using a process outlined in one of DeCODE's patent applications. Of course, if DeCODE's processes were any good, they would be Illumina, and they are definitely not. All DeCODE has is some samples and some statistical know-how, both of which are not patentable.
Wow! The Promethease site is really amazing. Thanks to all who posted their individual sequence info. Thanks for the links. Shouldn't a project like this deserve a government grant for putting together useful info from all the government funded gene research? I think so.
It always amazes me how no one wants to pay for all the work that goes into basic research and scientific discovery before the discovery is made and how many want it for free, after the discovery is made. It is one powerful reason why basic-research in the U.S. has literally died in the last quarter century.