Myriad, Mayo and Prometheus: the line between "law of nature" and invention

Some very interesting litigation has been going down this year, somewhat under the radar for most of us, regarding the patentability of biological products and processes that could have huge implications for both the biotech industry and academic research labs hoping to commercialize their findings.

Earlier this spring, what would look to most people (and even most scientists) as a very dry case made its way up to the Supreme Court. The case was brought by Mayo Clinic (via their laboratory group Mayo Collaborative Services) against Prometheus, a medical diagnostics company. Prometheus own(ed?)(s?) the patent on a certain test offered by the laboratory. This test involved detecting the levels of a certain drug metabolite in order to monitor dosage/predict side effects/etc., and correlate those levels to the health of the patient (i.e. efficacy). Basic pharmacokinetics/pharmacodynamics (PK/PD) type information--thousands upon thousands of people get a similar kind of testing for their coumadin/warfarin levels all the time, This is mainly because there are certain polymorphisms in a couple of enzymes (VKORC1 and CYP2C9) that affect the metabolism and thus blood levels of warfarin in some patients, and if the balance isn't right (and the effective dose ends up too high) the patient's blood won't be able to clot at all, which is a bad thing, obviously. If it's the other way around, the patient won't get enough of the thinning effect and will not get the intended benefit. So everyone who starts on warfarin has to get their prothrombin time and "international normalized ratio" (PT/INR i.e. measures of clotting) regularly checked to make sure the dose is having the intended effect; doctors adjust based on the outcomes of these tests as necessary.

So, Prometheus developed a test for a certain drug metabolite in a certain disease (unrelated to warfarin, that was just a handy example) that was a little more direct: it measured the actual level of the metabolite in the blood rather than a downstream effect of that level. (you could do this for warfarin too, but it's probably a lot cheaper to do the PT/INR test) They also established a correlation between those levels and the outcomes, and patented the whole system. They sold a kit, that made this test simple, to the Mayo labs, and Mayo Labs bought it and used it regularly. Then at some point, somebody at Mayo Labs said, "Hang on, if this is just measuring the level of this metabolite using a machine we already have, let's just put together our own reagents off the shelf and run it without paying all this money for this expensive kit." (something research labs may or may not do all the time... ahem...) As you might expect, Prometheus wasn't very happy about this and it all culminated in some intellectual property litigation. You know, just a little argument that ended up... before the Supreme Court.

Teh SCOTUS' decision ruled, in way-shortened summary, that because a drug metabolite is produced by the body, it is a "law of nature" and is thus unpatentable. This result sent waves through the intellectual property law community because if a diagnostic for drug levels is considered a "law of nature," WHAT ELSE is now going to be challenged as unpatentable?? The entire freaking biotech industry??!

Think, in particular, the ongoing fight about Myriad Genetics' test for BRCA1/BRCA2 mutations. Someone else certainly did. Last week, in an update to this law of nature drama, the U.S. Court of Appeals for the Federal Circuit ruled in favor of Myriad, that because the DNA constructs covered under their patents are cDNAs and other pieces of cloned out material, they are not "natural" and thus don't fall under the "law of nature" definition. This is a key result: it means that the issue isn't necessarily so sweeping and cut and dried as it looked from the opinion that came out of ye olde SCOTUS... and may ultimately end up back there via this Myriad case.

But overall, this whole discussion may end up having the effect that was speculated to be part of the SCOTUS' plan for their dramatic ruling: to better define what constitutes "natural" in the 21st century, where modern biotechnology can generate whole artificial genomes, and exome sequencing (heck, even the cheap kind of sequencing, or good old fashioned GC/MS metabolomics) can help predict who is at risk of a given disease and/or will benefit from a given drug; and to force the biotech world to figure out which and how molecular parts are owned by the individual who produced them (either in their body or from their mind). And us in academia, we better pay attention--because these are our discoveries out there, too, and if we want them to ever see the light of day, somebody is going to have to help us make products out of them. That's pretty hard to make happen without viable intellectual property in place.

8 thoughts on “Myriad, Mayo and Prometheus: the line between "law of nature" and invention

  1. Science tells us that "overconfidence bias" is something we all are prey to.

    To believe that these recent cases (Mayo and AMP) are about the "patentability" of the respective inventions is to fall prey to both our natural "overconfidence bias" and to the wordsmithing tactics of those who do not want us to understand what it is that is really being debated.

    It gets hyper-technical in the legal jargon of patents, but these cases are supposed to be strictly and only about section 101 of the US patent code (35 USC 101) and not about the complete set of tests (e.g. 102, 103, 112) for "patentability".

    If you are confused, do not feel bad. The Supreme Court is confused as well. (But they --in their beyond infinity wisdom-- are not about to admit it.)

    • First of all, what do you even mean by "overconfidence bias" in this context? That doesn't make any sense. That's really the only thing I'm confused about.

      Second, indeed, the "basic requirements" for obtaining a patent on something (i.e. patentability) do arise from section 101 (which simply states "Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title", and as interpreted by most courts, laws of nature and natural phenomena are not allowable as subjects of a patent (again, i.e. not patentable). The rest of the set of tests that you reference are part of the patentability test, but since the fundamental basic requirements must be first met RE: 101, it all points back to 101 being the first test of patentability. The other sections you cited provide much more detailed explication of what constitutes patentability, which make them easier to use for evaluation and prosecution, but they still build up from the base of 101 as what types of things can be patented.

      The biotech world has grown up in an age in which section 101 was largely glossed over by those granting and litigating patents; that is why the Supreme Court's citation of section 101 was such a big deal--everyone had been avoiding it for the most part because of the implications it would have for biotechnology IP and commercialization.

      Litigation of intellectual property issues, like all issues in law, is a matter of interpretation of these codes and rules by the lawyers, judges, courts to whom cases are brought. Interpretation is built on precedent--but precedent is not immune to counter by reinterpretation, that's how change rolls its iterative way through the legal system. So maybe you're confused about the role of the courts in defining and evolving the interpretations of the letters of the law? The Supreme Court might make decisions that not everyone agrees with, but they don't make them by accident because they are "confused."

      It would be surprising to me if the panel of intellectual property lawyers and the tech transfer experts with whom I've been discussing these things were "confused" too, you know, since this is the kind of thing they work with day in and day out.

  2. The understated part of this is the same issue at stake in software patent arguments: can you or can you not patent a mental process? What's not at issue is a novel arrantement of matter, but rather the inferences drawn from observing something -- in one of the given examples (BRCAx) a particular mutation, in another example an easily detected biochemical.

    In neither case is the method for detection being patented; both patents cover all possible methods of detection and in fact (IIRC) the Prometheus invention tests for an already known metabolite, although the test is more efficient than previous ones. What both Myriad and Prometheus are doing, however, is claim that the mental process of connecting the results of the test to clinical practice is itself practicing the patent.

    And that is the flip side of

    these are our discoveries out there, too, and if we want them to ever see the light of day, somebody is going to have to help us make products out of them.

    Because what happens to academics when someone can patent your mental processes?

    • Definitely--that's what makes all of this such a twisty issue for academics. For one our incentive structures don't necessarily reward this kind of thing, and for two the investment required to bring something to market is often beyond the resources and capabilities of ourselves (and in some cases even our tech transfer offices; some are outstanding, such as the one at my institution, but many others are lacking in expertise and institutional support).

      Prior art helps us in the mental processes category, since if our mental processes have been published or otherwise disclosed, they aren't patentable anymore (sometimes even by us if we don't get provisional patents submitted before we disclose them; that gets hairy). In the US we still have a year from first public disclosure to maintain our priority over something (i.e. presentation, paper, abstract, funded grant application which enters the public domain, or even discussion with outside people that isn't covered by a confidential disclosure agreement) for now, but that actually goes away as of some time next spring. So keeping up with the details on the value and protection of our intellectual property is pretty important, even though very few academics really pay much attention to it.

      • Just bear in mind that, as with software, the real problem is making sure that you aren't infringing someone else's patented thoughts by your work.

        • Well, for that we just have to get better at searching the patent databases as thoroughly as we do the research literature. (in general, academics are terrible about performing due diligence in the IP arena). A patent on something you're working on is the same as a previously published paper about something you're working on: it means you weren't the first person to think about/work on something, the previous work needs to be cited and usually will not incur prosecution if you're doing it solely for research purposes and not commercial purposes. Sometimes the owner of the patent will require you to work out a Materials Transfer Agreement with them, sometimes they'll just ignore your use of it.

          • What if the patent isn't on what you're working on, but the approach you're using to work on it?

            As far as searching patent databases, you might expect that with lotsabillions riding on it the people doing software would have found a way to avoid the problem. And, indeed, they have: patent language is so opaque that only a court can really determine whether you're infringing. However, if it turns up in discovery that you did the search you're on for willful infringement and the damages multiply.

            There's a reason every company I know of instructs its technical staff to never do any patent research. Which, if you think about it, is kinda backwards from the original idea of patents.

  3. Pingback: Links 8/26/12 | Mike the Mad Biologist

Leave a Reply to chemicalbilology Cancel reply

Your email address will not be published. Required fields are marked *