Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co. (Fed. Cir. 2010) (en banc)

    By Kevin E. Noonan --

It's settled, then -- unless the Supreme Court decides to weigh in, there is a separate written description requirement in 35 U.S.C. § 112, first paragraph.  The precedent developed since Regents of the University of California v. Eli Lilly & Co. is good law.  This is the judgment of the Federal Circuit in an opinion by the en banc Court, written (appropriately) by Judge Lourie, and joined by every member of the Court except (predictably) Judge Linn and soon-to-be Chief Judge Rader.

The en banc opinion addressed all of Ariad's arguments, and found in favor of Lilly.  Ariad helped the Court in arriving at this decision, in how it framed its responses to the questions presented by acknowledging that the statute contained a requirement for a written description.  Ariad's position, according to the opinion, was that the measure of whether the written description was satisfied was enablement, and here the Court parted ways with Ariad's argument.  The Court characterized this situation as the parties agreeing that the specification must contain a written description of the invention, but disagreeing on the standard for fulfilling it (as well as whether the requirement applies to originally-filed claims).  The Court discussed Ariad's grammatical construction of 35 U.S.C. § 112, first paragraph, and particularly the argument regarding the significance of the comma after the phrase "and the manner and process of making and using it," and decided that Lilly's interpretation is the correct one.  Specifically, the Court held that § 112, first paragraph, contains two separate description requirements:  a written description of the invention and a written description of how to make and use it.  The Court dismissed the grammatical argument thusly:

[W]e see nothing in the statute's language or grammar that unambiguously dictates that the adequacy of the "written description of the invention" must be determined solely by whether that description identifies the invention so as to enable one of skill in the art to make and use it.

The argument rests on the parallelism of the language (an analysis that forms one of the bases for Judge Rader's dissent, joined by Judge Linn).  The Court also dismissed Judge Linn's position, that enablement is sufficient to satisfy the requirement of § 112, first paragraph, but on the backs of several of the amici's arguments (Judge Linn makes his "enablement is enough" argument in a separate dissent, joined by Judge Rader).  According to the en banc opinion, this reading of the statute renders either "and the manner and process of making and using it" or "[a written description] of the invention" as surplusage, "violating the rule of statutory construction that Congress does not use unnecessary words," citing U.S. v. Menasche, 348 U.S. 528, 538-39 (1955).  The Court completed this portion of its analysis by citing previous versions of the Patent Act, as well as first principles:  "[e]very patent must describe an invention.  It is part of the quid pro quo of a patent; one describes an invention, and, if the law's other requirements are met, one obtains a patent."

The en banc Court then reviewed the several instances of Supreme Court precedent raised by the parties, including Evans v. Eaton, O'Reilly v. Morse, Schriber-Schroth Co. v. Cleveland Trust Co., and Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., agreeing with Lilly that this precedent recognizes a written description requirement that is separate from the enablement requirement of § 112, first paragraph.  And the Court rejected the argument that interpreting § 112, first paragraph, as containing a separate written description requirement conflicts with § 112, second paragraph, saying that the claims "define the subject matter that, after examination, has been found to meet the statutory requirements for patent"; their purpose is not to describe the invention, but "to provide notice of the boundaries of the right to exclude and to define [the] limits" of the invention.

The Court also held that stare decisis principles required its decision.  Having expounded on the existence of a separate written description requirement for over forty years, the Court held that they must "tread lightly" in upsetting the "settled expectations" of the patenting community.  "If the law of written description is to be changed, contrary to sound policy and the uniform holdings of this court, the settled expectations of the inventing and investing communities, and PTO practice, such a decision would require good reason and would rest with Congress."  The Court also opined that, while its use of the term "possession" regarding the written description requirement has "never been very enlightening," at the same time the Court had consistently applied the law in this area.  And any inconsistencies in the application of the law in this area arose, according to the Court, because satisfaction of the written description requirement is a question of fact, and the different facts in each case (and arguments based on those facts) produced (not unexpectedly) varied conclusions.  Thus, the Court did not intend to attempt to set forth any "bright line rules," for example, with regard to the number of species required to adequately describe a genus (thus dashing any hopes that this particular area of uncertainty would be settled by this opinion).  However, there were "a few broad principles" that the Court said "hold across all cases":

We have made clear that the written description requirement does not demand either examples or an actual reduction to practice; a constructive reduction to practice that in a definite way identifies the claimed invention can satisfy the written description requirement.  Falko-Gunter Falkner v. Inglis, 448 F.3d 1357, 1366-67 (Fed. Cir. 2006).  Conversely, we have repeatedly stated that actual "possession" or reduction to practice outside of the specification is not enough.  Rather, as stated above, it is the specification itself that must demonstrate possession.  And while the description requirement does not demand any particular form of disclosure, Carnegie Mellon Univ. v. Hoffmann-La Roche Inc., 541 F.3d 1115, 1122 (Fed. Cir. 2008), or that the specification recite the claimed invention in haec verba, a description that merely renders the invention obvious does not satisfy the requirement, Lockwood v. Am. Airlines, 107 F.3d 1565, 1571-72 (Fed. Cir. 1997).

The majority opined that while there may be many instances where there is "little difference" between describing an invention and enabling it, that will not always be the case, particularly in the chemical and biotechnological arts.  A separate written description requirement "plays a vital role in curtailing claims that that do not require undue experimentation to make and use, and thus satisfy enablement, but that have not been invented, and thus cannot be described."  And "particularly for the biological arts," having a separate written description requirement "ensures that when a patent claims a genus by its function or result, the specification recites sufficient materials to accomplish that function."

Turning to the distinction, if any, between amended claims and original claims, the Court held that the analytical differences espoused by the parties with regard to the case law concerning amended claims -- Ariad contending that the distinction was based on enablement and Lilly on written description -- were semantic differences that did not change the outcome.  Regarding original claims, the Court agreed with Lilly that, although rare, even original claims could encompass subject matter not sufficiently described to satisfy the disclosure requirements of § 112, first paragraph.  There is "no principled basis" for distinguishing the requirements of original claims versus amended claims, or for restricting the written description requirement to policing priority (including nothing in the statutory language to support such a limitation).  The "truism" that the claims are part of the original specification, as advocated by Ariad, does not mean that the invention has been sufficiently described merely by reciting it in an original claim:

For example, a generic claim may define the boundaries of a vast genus of chemical compounds, and yet the question may still remain whether the specification, including original claim language, demonstrates that the applicant has invented species sufficient to support a claim to a genus.  The problem is especially acute with genus claims that use functional language to define the boundaries of a claimed genus.  In such a case, the functional claim may simply claim a desired result, and may do so without describing species that achieve that result.  But the specification must demonstrate that the applicant has made a generic invention that achieves the claimed result and do so by showing that the applicant has invented species sufficient to support a claim to the functionally-defined genus.

"[G]eneric claim language appearing in ipsis verbis in the original specification does not satisfy the written description requirement if it fails to support the scope of the genus claimed."  The Court also rejected Ariad's argument that the written description requirement was disproportionately applied to biotechnology inventions:  generic language in an application as filed does not automatically satisfy the written description requirement in instances such as Fiers v. Revel, Eli Lilly and Enzo v. Genprobe.

The majority also rejected Ariad's argument that the Court applied the written description requirement disproportionately to basic research and inventions by academics and universities.  This result is not inconsistent with sound policy that "basic" research may not lead to patentable inventions, characterizing the difference between "academic theories, no matter how groundbreaking," which are not patentable with inventions having a practical use, citing Brenner v. Manson that "[a] patent is not a hunting license.  It is not a reward for the search, but compensation for a successful conclusion."  While this situation may decrease incentives for university research, the Court is mindful of disincentives on "downstream research"; the goal, according to the Court, is to strike the right balance, which the majority believes it does:  "giving the incentive to actual invention and not 'attempt[s] to preempt the future before it has arrived,'" citing Fiers.

Turning to the subject matter of the appeal, the Federal Circuit reversed the District Court's denial of JMOL, on the basis that no reasonable jury could have found for Ariad based on the facts adduced at trial.  Specifically, the CAFC held that Ariad's evidence was either directed to the wrong priority date (the jury having found that the patent-in-suit was entitled to a priority date in 1989 while the majority of the evidence did not extend earlier than 1991) or was insufficient to support the "vast scope of these generic claims," especially in view of the "gaping holes in its disclosure" and the "primitive and uncertain" state of the art (considerations that, admittedly, resonate with the Wands factors used to assess compliance with the enablement requirement).  The Court specifically disapproved Ariad's strategy of avoiding reciting any NF-kB inhibitor molecules in the claims in an effort to avoid the court's holding in University of Rochester v. G.D. Searle (that claims were invalid for failure to satisfy the written description requirement based on the failure to support a claim term for an inhibitor of COX-2).

Judges Newman and Gajarsa filed separate concurring opinions, Judge Newman addressing the need to protect early-stage technologies such as those developed by universities, and Judge Gajarsa opining that the majority's statutory construction was "reasonable" but not the only (or even a preferable or necessary) way to construe the statutory disclosure requirements (calling the text of 35 U.S.C. § 112, first paragraph, "a model of legislative ambiguity").  Judge Linn dissented based on his opinion that enablement is a sufficient measure of the proper extent of disclosure, and Judge Rader dissented (vigorously) based on his entirely contrary interpretation of the statute.

Regardless of these disparate voices, the Court majority affirmed the existence of a separate written description requirement.  This opinion puts to bed one of the many recurring examples of uncertainty that have crept into its jurisprudence in recent years (at least until sufficient members of the Court have accepted senior status and new judges bring their views to bear on the question).

Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co. (Fed. Cir. 2010) (en banc)
Before: Chief Judge Michel and Circuit Judges Newman, Mayer, Lourie, Rader, Bryson, Gajarsa, Linn, Dyk, Prost, and Moore
Opinion by Circuit Judge Lourie, joined by Chief Judge Michel and Circuit Judges Newman, Mayer, Bryson, Gajarsa, Dyk, Prost, and Moore; additional views by Circuit Judge Newman; concurring opinion by Circuit Judge Gajarsa; dissenting-in-part and concurring-in-part opinion by Circuit Judge Rader, joined by Circuit Judge Linn; dissenting-in-part and concurring-in-part opinion by Circuit Judge Linn, joined by Circuit Judge Rader

Comments

[KB]

Perhaps someone who has followed this case since the beginning could enlighten me: Couldn't all or most of the claims in the Ariad patent have been found invalid as inherently anticipated? Don't many old and well-known drugs act by the claimed mechanisms?

[EG]

Kevin,

No surprises here, incluidng that lack of a unanimous opinion or that Lourie would write the majority opinion or that Rader would dissent. In my opinion, Newman's concurrence concisely summarizes the essence of what the first paragaph of 35 USC 112 is about, better and in fewer words than Lourie.

[Andrew Williams]

Dear KB,

The issue of inherent anticipation was tried before a jury during the proceedings at the District Court level. As you are probably aware, the concept of inherent anticipation is not an easy one for most people, and especially for individuals with no (or minimal) legal or scientific training. Correspondingly, the claims at issue were found by the jury to not be anticipated. However, the 6,410,516 patent is currently the subject of Reexam No. 90/007,503 (the image file wrapper for which is available at the patent office website). Some of the claims of the '516 patent have been finally rejected at least because they are inherently anticipated by the use or administration of the following compounds or drugs: protein kinase C inhibitors, cyclosporin A, calcitriol (the active form of vitamin D), 5-ASA, glucocorticoids, antibiotics, and N-Acetyl-L-Cysteine. An appeal brief has been filed by the patent holders.

[6]

A good write up. Anyone you guys know of up in arms in the university circles?

[Kevin E. Noonan]

Dear 6:

The decision preserves the status quo, so while many in the biotech space think it is wrong, they have had 14 years to adapt their practices to the law as it has been and now remains.

On the bright side, the decision certainly points to the benefits of a greater town/gown collaboration between pharma and universities and their startups, something that has long been a tradition in other areas (petroleum and chemical industries partnering with chemical engineering faculty, for example). The biggest remaining problems stem from the different timelines between academic publication schedules and patenting; unlike for companies, professors cannot wait years before disclosing their experimental results (nor would that be a good outcome for society).

Perhaps there needs to be less emphasis on targets (which were the inventions du jour from the Human Genome Project) and more on compounds (small molecules or biologics) that interact with targets to comprise useful drugs. That seems to be where Judge Lourie and his brethren were going in the part of the opinion dealing with the impact (negative, disparate) of a separate written description requirement on universities.

One thing that is certain: if there was ever a group that would continue to "innovate" regardless of the existence of patent protection, it should be university professors. In this context, the benefits of patenting are to attract investment necessary for commercialization. That effort (whether for diagnostic or therapeutic applications of basic research) is many times more costly and uncertain than the original research, and needs patent protection for investors to take the risks associated with early-stage biotechnology companies.

Thanks for the comment.

[EG]

Kevin,

As I mentioned what Ariad held isn't surprising or even new to me, a patent attorney with a chemical background who has had to deal with the the “supporting the scope of the genus” problem is longstanding issue with us. I’ll also agree that Judge Lourie is not the clearest on what satisfies the “written description” requirement (and his opinion in Prometheus on when a drug dosage calibration method satisifes Bilski is a model of obfuscation of the point).

My biggest problem with Lourie’s opinion is that it spends an inordinate amount of time on the “separate and distinct” aspect of the “written description” requirement which I characterize in my article as equivalent to “how many angels can stand on the head of a pin.” As far as I was concerned, this question was settled by the Federal Circuit at least as far back as the 1996 case of In re Alton, and even earlier by the Supreme Court.

What I was more concerned about Ariad resolving is what does this "written description" requirement (be it "separate and distinct" or otherwise from the "enablement" requirement) mean. Probably the most significant statement in this regard is at page 21 of Lourie's opinion is where he says (referring to Univ. of California v. Eli Lilly) "an adequate written description of a claimed genus requires more than a generic statement of an invention’s boundaries." I call this the "black box" analogy because you may define the boundaries of a "black box" without ever defining what its contents are. What Lourie is looking for (and which makes sense to me) is some definition of what's in the "black box," or in patent claim lingo, what's in the broad genus, functionally, structurally, etc. What the ‘516 patent did in Ariad (as the Young patent in the Rochester case also unfortunately did) was simply define the boundaries of the generic “black box” but didn’t give you sufficient explanation as to how define and identify what molecules (in sufficient quantity) that would fit within that ratther large generic “black box.”

I've applauded Newman’s concurrence because it concisely distills what the “essence” of the first paragraph of 35 USC 112 is: “the patentee is obliged to describe and to enable subject matter commensurate with the scope of the exclusionary right.” That’s the key: if you want broad generic coverage, you must adequately describe and enable that scope in the patent spec. As I noted in my article, that’s been true since 1853 O’Reilly v. Morse case (cited directly by Newman and unfortunately only addressed in a footnote by Lourie) which has regretably been miscited (repeatedly) as a 35 USC 101 case, but is really a first paragraph 35 USC 112 case.

My additional 2 cents for what it’s worth.

[Derek]

I'm with Judge Newman: "the real issue of this case is too important to be submerged in rhetoric".
I see Ariad and Rochester as being similar to the "reach-through claim" cases, where the discoverer of a phenomenon attempts to claim (read, receive money from) all yet-to-be-made downstream inventions flowing from the discovery:- in the pharma world, think "A therapeutic compound found by the method of claim 1". Yet interestingly the stake was driven through the heart of reach-through claims in the US, and it seems also in Europe, without all the fuss engendered here.
Perhaps Ariad will live on as one of the "right result, wrong reasoning" cases, perhaps it will come to be considered as "right result, right reasoning", but I'm definitely in the "right result" camp as far as how it came out.

[Kevin E. Noonan]

Dear EG and Derek:

I think the point is there cannot be any hard and fast rule, since what is enough disclosure to satisfy the requirement will depend on the facts in each case. Now, it does create uncertainty, much like the question of obviousness, since it may not be a simple matter to have the scope of the claims conform to the extent of the disclosure. The remedy, as imperfect as it may be, is to write claims of varying scope. If that leaves room for a third party to design around, the answer is to make the design around difficult, unpredictable, expensive and inferior to the claimed invention.

Sometimes a tall order, but the alternative are overbroad claims that prevent innovation. The goal of patent law to promote the "better" mousetrap recognizes that there will be an original mousetrap that is the basis of the improvement.

Thanks for the comments.

[EG]

"but the alternative are overbroad claims that prevent innovation."

Kevin,

I view "overbroad claims" somewhat differently: if the patent spec doesn't describe or enable that broad claim scope, the inventor fails on his part of the "quid pro quo" extracted by the first paragraph of 35 USC 112. That's been true at least as far back as O'Reilly v. Morse.

I also completely agree with your statement that there can't be "any hard and fast rule" on what is adequate description and enablement for broad generic scope claims. The problem of supporting a broad generic scope claim has been with us since I started practice back in 1977 and before. Doing this in the newer "unpredictable" biotech is just another variant on this old problem. As my patent attorney brother Mark and I discussed, there's really no "magic number" for the number of species to support the genus. It all depends on the technology that is claimed. Certainly having a number of species identified, along with a descriptive formula for how to identify others will usually do it.

What academic research unfortunately tends to do is discover the scientific mechanism and stop there. That's what Judge Newman says won't get you a valid broad generic scope claim. Instead, you've got to do is describe and enable (somehow) what's in that broad generic "black box" in a way that the PHOSITA can predict (beforehand) what compounds will satisfy that mechanism. Academia can complain all they want to about the "unfairness" of this rule, but, like the narrow "experimental use" defense, the courts are not persuaded by those complaints.

If you put the holdings and fact patterns together from Univ. of California, Univ. of Rochester, Ariad, and the 2004 In re Wallach cases together, you do start to get a picture of what will satisfy "written description," along with "enablement." (I'm also in Newman's camp that you need to approach "written description" and "enablement" in tandem, not separately). What would really help, and what we currently don't have, is a decision involving a broad generic biotech claim where the Federal Circuit confirms that "written description (as well as "enablement") were satisfied. When that happens, we will at least know one way to reach the goal.

[Kevin E. Noonan]

Of course, EG, then there are the instances where an inventor can enable an invention without describing it (think California v. Lilly). That's where the problems arise.

Thanks for the comment.

[EG]

Kevin,

Yep, that "black box" called human insulin cDNA. Doesn't tell you what nucleotides (and in what order) that cDNA comprises.