Well now, isn't this an interesting development. As covered by many news outlets (NYT, Wired, Genomeweb), US District Court Judge Robert Sweet has invalidated several US patents, sometimes referred to as the "BRCA1/2 patents", held by the University of Utah and Myriad Genetics. From Judge Sweet's decision: "Products of nature do not constitute patentable subject matter absent achange that results in the creation of a fundamentally new product." Judge Sweet's decision is here (PDF)
via Genomics Law Report. Here is the ACLU's take.
Here is a brief summary of what follows: The ruling is remarkable. Various commentators and reporters remark upon it. They get confused. I try to clarify. Then we get to a truly revolutionary part of the decision: it's about science! And a little bit about law. Finally: so what if a few patents are invalidated?
Didn't See That Coming. But I Can't Complain.
Last month, I noted that I was skeptical that the ACLU and other plaintiffs would be so successful in one go. So I am surprised, but I am certainly not disappointed. But I am not surprised, while being somewhat disappointed, that the coverage of the decision is so confused and confusing. This confusion arises, I suspect, because the wording of Judge Sweet's decision is not entirely straightforward in places, and this has led to analyses that are insufficiently careful. More on these points below.
DISCLAIMER: Please recall in what follows that I am but a humble physicist by training (oh yes, yes, we're all very humble), not a lawyer. But I have written some stuff about patents on genes, and at least a few people (some of whom are IP law lawyers) think my analysis doesn't suck a lot.
First, over at Genomics Law Report (GLR), John Conley and Dan Vorhaus have a great analysis with a nice title: "Pigs Fly: Federal Court Invalidates Myriad's Patent Claims". I won't bother to repeat their discussion. If you are interested in this issue, please read that post as well as Dan Vorhaus' initial post analyzing the decision. In particular, the reader might want to attend closely Vorhaus and Conley's observations about the potential for appeals, the likelihood of success in that endeavor, and the applicability of the ruling in other jurisdictions.
The short summary of what's transpired so far in the case is that Judge Sweet has invalidated a small number of claims, in a summary judgement ruling that so far applies only in the Southern District of New York. Assertions that this is the end of the world for companies that hold gene patents are rather overblown.
There's Too Much Confusion, But Here is Some Relief
But now onto some of the confusing bits alluded to above. The confusion starts, surprisingly, at GLR. Here are Conely and Vorhaus: "In the broader policy debate surrounding gene and biotechnology patents, however, this decision is the latest, unmistakable shot across the bow of gene patent holders, particularly those such as Myriad Genetics that have developed businesses around patent-protected genetic tests supported by exclusive rights in underlying gene patents." Hummm... Maybe not so much, actually. Let me get straight to the point: there is a rather substantial difference between a "gene patent" that claims naturally occurring sequences and one that claims sequences that are not natural.
Here is one way to think about the issues under discussion: in my one hand, I have a piece of isolated DNA that is identical in sequence to one in your body. It is the same genetic sequence, so it carries the same information. Indeed, for it to be useful in a test tube for the purposes of diagnosis, it must have both the same information content and the same function as the sequence in your body. In fact, it only works as a diagnostic tool because it is the same as what is in your body. As I noted in my earlier post, this is sort of the opposite of invention, and I have never understood why natural genes can be patented. (Note: Judge Sweet hits this point quite squarely, but not until p.124 of his ruling.) In my other hand, I have a piece of isolated DNA that is solely the result of human manipulation -- "human ingenuity" -- consisting of a sequence that does not exist in nature. Both pieces of DNA are isolated, but they derive from very different sources, and are derived by very different means. Unfortunately, everybody discussing the present decision, including Judge Sweet in the early pages of his decision, seems to be a tad careless about the distinction, which leads many people down a rabbit hole. (There is an extended discussion of the definition of "isolated DNA" and of the BRCA1/2 genes on p.90-92.)
Here is where it starts: Judge Sweet sets up his decision in the first couple of pages focusing specifically on the BRCA1/2 genes, and slightly more generally on isolated human genes: "Are isolated human genes and the comparison of their sequences patentable?" (p.2) He continues: "Two complicated areas of science and law are involved: molecular biology and patent law. The task is to seek the governing principles in each and to determine the essential elements of the claimed biological compositions and processes and their relationships to the laws of nature."
This sounds great. Judge Sweet is clearly referring specifically to certain human gene sequences named in the patents in question. Alas, on the next page he switches his language to address the specific assertions of the plaintiffs that ""isolated DNA" containing human BRCA1/2 sequences" are not patentable. The basic contention here is that because the isolated DNA as described in the patents does the same thing inside the body as outside the body -- it is an information storage medium -- there is no difference between the two forms of DNA and therefore the isolated DNA in question cannot be patented. Judge Sweet concludes (p.4):
DNA represents the physical embodiment of biological information,
distinct in its essential characteristics from any other chemical found
in nature. It is concluded that DNA's existence in an 'isolated' form
alters neither this fundamental quality as it exists in the body not the
information it encodes. Therefore, the patents at issue directed to "isolated DNA" containing sequences found in nature are unsustainable as a matter of law and are deemed unpatentable subject matter.
The judge thereby switches within a couple of paragraphs very seamlessly from language referring only to human genes to language referring seemingly to all "isolated DNA". It takes another 100 pages to get to a true clarification, and I'll bet very few people have read that far, or followed all the byways and cross-references (p.100): "...The issue presented by the instant motions with respect to the composition claims is whether or not claims directed to isolated DNA containing naturally-occurring human sequences [emph added] fall within the products nature exception. ...It is concluded that the composition claims-in-suit are excepted."
In other words, Judge Sweet very specifically ruled that the claims on isolated DNA containing naturally occurring sequences are not valid. Even more specifically, the ruling only applies to the motion in question by the plaintiffs, namely to invalidate the patents on BRCA1/2 held by Myriad et al. Judge Sweet pointedly cites Diamond vs. Chakrabarty
(p.109) -- a case that affirmed the patentability of "genetically
engineered" organisms -- in limiting his ruling to the patentability
of naturally occurring genes. The ruling has no applicability outside that subject matter, and therefore has little applicability to, for example, much of anything that might come out of synthetic biology (unless you are talking about a synthetic DNA version of a naturally occurring gene). Nor, for that matter, does the ruling have any say about any bit of DNA altered to be different from a natural sequence. Which means that the ruling has very little to do with most patents on DNA, and therefore has very little to do with most of the industry surrounding those patents -- more on this below.
(Side note, as I read through the decision: Myriad's lawyers didn't do themselves any favors by making generally unpersuasive assertions aimed as broadside attacks against the plaintiffs' arguments. As noted in my previous post on this case "Whither Genome Patents?", the defendants' assertions that patents serve as necessary incentives for scientific research are complete bunk. Defense attorney Brian Poissant previously argued that "women would not even know
they had BRCA gene if it weren't discovered"
under a system that incentivizes patents. I say again, as calmly as I can, bull pucky. For example, see the publicly funded Human Genome Project. See also the fact that BRCA2 was sequenced first in academic labs rather than by Myriad, who somehow managed to patent it anyway. See also the many BRCA1/2 assays independently developed in academia, the use of which Myriad repeatedly quashed through cease-and-desist letters, as recounted in detail in the decision. But here is Judge Sweet himself (p.76): "According to Myriad, its policy and practice has been and still is to allow scientists to conduct research studies on BRCA 1 and BRCA 2 freely, the result of which has been the publication of [over 8600 papers] representing the work of over 18,000 scientists." (It wasn't clear to me whether Myriad's legal team itself provided these numbers -- but if they did: bad legal tactics, fellas.) In other words, 18,000 scientists have managed to produce a substantial body of work without any promise whatsoever of remuneration based on a patent for BRCA1/2. Unless, of course, you count keeping your job through the promise of not being sued by Myriad.)
It's Science! And Science Always Wins -- Eventually, But May be Delayed By Appeals.
There is another very interesting angle to Judge Sweet's decision. Andrew Pollack, writing in the New York Times, suggests that the most revolutionary part of the decision is where Judge Sweet recognizes that DNA carries information. Pollack quotes Rebecca Eisenberg, a law professor at the University of Michigan: "There isn't a whole lot of doctrinal support" for considering DNA as information rather than as a chemical. That, for me, is a truly eye opening perspective. Not because I didn't know about it before -- unfortunately, that view is all too prevalent among IP lawyers -- but rather because it is being defended and suggested as a possible grounds for appeal. True, it may be precedent, but that does not mean it is good precedent.
Here's the thing: There may not be much "doctrinal support" for considering DNA as information, but there is a rather overwhelming amount of scientific and technical support for considering DNA as information rather than as a chemical, say starting with the vast majority of molecular biology and biochemistry papers published in Science, Nature, Cell, PNAS, and any other relevant journal you can think of. For all of the last six decades, no less. Oh, and then all those silly textbooks. The genetics and molecular biology ones, obviously; not the law textbooks.
Judge Sweet, in my humble opinion, already smacked this one out of the park on p.4: "The facts relating to molecular biology are fundamental to the patents at issue and to the conclusions reached. Consequently, in the findings which follow, the discussion of molecular biology precedes the facts concerning the development, application, and description of the patents." (Whoa there! Science and reason trump the law of man! Or science and reason trump the law of lawyers? Damn, now that is a novel legal theory. And a welcome one. Don't tell Sen. James Inhofe.)
Unfortunately, Pollack misses this angle, and promulgates further the confusion that Judge Sweet's ruling spells doom for the biotech industry: "Some biotechnology investors and executives say that lack of patent protection for DNA could diminish investment in the field and remove incentives for companies to develop tests." Never mind that, as described above, Judge Sweet's ruling applies only to patents on naturally occurring genes, which should ameliorate the concerns of most of the "some biotechnology investors and executives". It is nonetheless true that diagnostics companies that rely on patents claiming naturally occurring sequences may have to reevaluate their business plans. (For instance, they may want to be especially careful in issuing cease-and-desist letters, lest the ACLU and company get busy again.) And it may be true that this small fraction of biotech businesses may have difficulty raising capital -- but time will tell. If it turns out that development of new diagnostic assays lags as a result of more patents on human genes being invalidated, then we will have something real to talk about. We might consider developing public policy around alternate incentives. Until there is a demonstrated concern, however, it isn't clear to me that we should be so concerned about the fate of private investors who gambled on patents whose validity has long been questioned.
What Is The Real Impact Going To Be?
To reiterate the numbers from my earlier post: of the roughtly 2% of US GDP that is derived from biotech, at a rough guess I would put only 1% of the total (so .01% of US GDP) in the molecular diagnostics category that depends explicitly on excluding other uses of patented human genes. A few billion dollars a year, in other words, might be at risk. But somebody is going to do the tests, and Judge Sweet's decision lists a variety of tests that cost about 1/3 of Myriad's; that is, before Myriad shut them down with cease-and-desist letters. If you eliminate those patents, we might have to come up with some other way to incentivize the development and testing of assays. Prizes come to mind as a fine thing to try. They work. Academics and garagistas will be happy to compete for those prizes, I am sure.
But the rest of the biotech industry shouldn't be concerned about this ruling, frankly. They might even celebrate the fact that they now have access, potentially, to a whole bunch more genes that are naturally occurring. Not just in humans, mind you, but any organism. This opens up a rather substantial toolbox for anybody interested in using biological technologies derived from viruses, bacteria, plants, etc. If it holds up over the long run, Judge Sweet's decision should accelerate innovation. That is definitely a good thing.