by Adriana Benedict
Today, Professor Glenn Cohen announced on this blog that he, in conjunction with two others, filed an amicus brief in AMP v. USPTO (Myriad), a case concerning Myriad’s patents on isolated DNA and cDNA. In a paper I have been writing on the natural phenomenon doctrine as applied to biotechnology patents, I arrived at this conclusion about the doctrine’s implications for Myriad:
According to Mayo v. Prometheus, the preemption rationale for the natural phenomenon doctrine suggests that any patent on a diagnostic biotechnology product or process should be limited to the inventive use of that product or process as defined by its associated process or product, respectively. As applied to Myriad, this qualified interpretation of the natural phenomenon doctrine would suggest that ideally these patents ought to be limited to Myriad’s one remaining valid method claim, namely claim 20 of the ‘282 patent, “a method for screening potential cancer therapeutics.” The unavoidable and unsettling problem with such a conclusion, of course, is that at this stage in litigation, it is not possible for the Court to limit Myriad’s gene patents in this way. This procedural limitation sheds some light on the elephant in the natural phenomenon doctrine: If the doctrine was meant to exclude certain categories of discoveries from patentability before Congress had the opportunity to refine more specific patent validity rules, then perhaps it should be limited to carrying out that function at the outset of a patent prosecution. The natural phenomenon doctrine serves the important purpose of ensuring that patents do not contravene their Constitutional objective by too broadly preempting the use of “basic tools of science.” It does so by balancing the scope of preemption against the scope of invention, and ensuring that the scope of preemption does not exceed that which is justified by the inventor’s handiwork in applying natural phenomena. At the patent prosecution stage, the natural phenomenon doctrine is a useful “catch-all” analytical tool that allows flexibility in promoting the spirit of patent law when the letter of patent law has not kept pace with the progress of science. But at the litigation stage, its Achilles heel is that it may prove too much: In the absence of a procedural option to limit a patent at this stage, the natural phenomenon doctrine is forced to err on either the side of all or nothing. While the doctrine may be useful at the patent prosecution stage, it was not (as other statutory patentability requirements were) appropriately designed to assess the validity of patents once they’ve been issued in a way that is compatible with today’s patent litigation procedures. As a doctrine of limitation, it must in this context either fall, and prove nothing at the expense of unwarranted preemption, or rise, and prove too much at the expense of patent holders who have been reasonably relying on guidance from the USPTO regarding gene patents for many years.
I am unable to find any commentary exactly on this point, but some issues concerning the jurisdictional authority of §101 have been raised in response to both Mayo and CLS Bank v. Alice. While these cases concern biotechnology processes and software, respectively, they are extremely relevant to Myriad if we consider isolated genes / cDNA to be the equivalent of biological software. Indeed, Professor Ronald Mann observed that “Though most of the attention to …[Mayo] has focused on its immediate implications for medical providers, the broader effect of the case probably will be on the software industry.”
Among the most striking of these is Professor David Hricik’s argument that §101 non-patentable subject matter (including judicial exceptions like the natural phenomenon doctrine) may not be grounds for invalidity as a defense to infringement, because §282 (presumption of validity; defenses) does not specify it as such, despite a seemingly exhaustive list of options. As Professor Hricik points out, §282’s inclusion of “any ground specified in part II as a condition for patentability” should extend defenses only based on §§102 and 103, which are clearly titled “Conditions for Patentability.” Professor Mark Lemley has offered a possible response to this that perhaps “101 are threshold issues that are not defenses at all, but pure questions of law, and therefore not things on which the defendant bears the burden of proof, and which can be raised even if not pled . . . at any time, though it would be a departure from current Federal Circuit precedent.” But the Supreme Court’s post-cert grant dismissal of LabCorp v. Metabolite as improvidently granted, presumably because the issue of §101 eligibility was not argued below, would run counter to this conclusion. On the other hand, Professor Lemley’s conclusion finds implicit support in Justice Breyer’s dissent to the dismissal. If Justice Breyer had it right, then Professor Lemley is correct that the issue is settled, but we cannot (at least not yet) accept this is as law. Professor Dennis Crouch presents another level of complexity added by the reformulation of §282 in the America Invents Act, which now applies also to defenses for “post-grant review proceedings directed to finance related business method patents as part of the Transitional Program for Covered Business Method Patents.” Professor Crouch points out that the USPTO’s final rules permit §101 challenges. Yet it is not entirely clear that the USPTO’s interpretation of §282 truly resolves this question. In fact, the final rules explicitly note that:
a party may be authorized to file a single motion for unpatentability based on prior art, a single motion for unpatentability based upon failure to comply with 35 U.S.C. 112, lack of written description, and/or enablement, and potentially another motion for lack of compliance with 35 U.S.C. 101, although a 35 U.S.C. 101 motion may be required to be combined with the 35 U.S.C. 112 motion (my emphasis added).
It’s curious why the USPTO would speak so ambivalently about “potential” §101 challenges if there wasn’t a jurisdictional predicament lurking in the background.
Professor Crouch also points out that these issues were raised in CLS Bank’s brief for the en banc rehearing of CLS Bank. He notes, nevertheless, that CLS Bank may have taken the argument too far by suggesting that eligibility and validity are two separate inquiries, and there ought to be no presumption of validity in determining subject-matter eligibility; Mayo’s conclusion of invalidity (rather than “ineligibility”) does not appear to support such an approach. To some extent, though, this is getting into semantics, as the Court in Mayo clearly found the patent at issue to be directed at ineligible subject matter.
So what’s a court to do? It seems most likely that they will continue glossing over the fact that Congress excluded subject matter eligibility from §282. In the alternative, any meaningful consideration of this dilemma could result in broad havoc. Time to start planning for a “Transitional Program for New Technologies Raising Subject-Matter Eligibility Issues”?
