Idenix Pharmaceuticals LLC v. Gilead Sciences Inc. (Fed. Cir. 2019)

By Kevin E. Noonan --

Section 112 of the Patent Act as codified, entitled "Specification" in the statute, specifies the amount of disclosure required to support a patent claim (among other requirements).  Section 112(a) contains three requirements:  written description, enablement, and best mode (although the latter has been in something of a state of limbo since the Leahy-Smith America Invents Act (AIA) disabled it as a defense).  Today in Idenix Pharmaceuticals LLC v. Gilead Sciences Inc., the Federal Circuit held that Idenix's patent was invalid on both grounds, affirming the District Court's overturning of a jury verdict on enablement and the District Court's post-trial denial of judgment as a matter of law (JMOL) regarding satisfaction of the written description requirement.  In doing so, the Court illustrated ways in which it has been able to impose its views (and recently, the Chief Judge's views) on both aspects of Section 112 requirements despite its reliance on fact finding by the jury or district court below (with Judge Newman characteristically dissenting from what she viewed as appellate court overreach by her brethren).

The case arose in litigation over Idenix's U.S. Patent No. 7,608,597 that was directed to drugs for treating hepatitis C virus (HCV), which Idenix alleged Gilead would infringe by launch of its sofobuvir (Solvadi®) HCV treatment.  Independent claim 1 of the '759 is representative of Idenix's invention:

1.  A method for the treatment of a hepatitis C virus infection, comprising administering an effective amount of a purine or pyrimidine β-D-2'-methyl-ribofuranosyl nucleoside or a phosphate thereof, or a pharmaceutically acceptable salt or ester thereof.

The opinion illustrates the structure of the purine or pyrimidine β-D-2'-methyl-ribofuranosyl nucleoside as disclosed in the '759 patent:

which differs from naturally occurring embodiments by the substitution of a methyl group at the 2' position on the ribofuranosyl sugar, cis to the nitrogenous base (or in the "up" position as understood by the Federal Circuit).  Gilead argued (and the District Court and Federal Circuit agreed) that the '759 specification did not provide guidance regarding the "billions" of possible molecules falling within the scope of the claims.  This argument was based on the acknowledged difference between the compounds exemplified in the '759 patent (having a hydroxyl, -OH, group at the 2' "down" position) while Gilead's accused infringing compound had a fluorine atom at that position.  After protracted ("years," according to the opinion) litigation, the District Court conducted a jury trial in which Gilead conceded infringement but challenged the '759 claims as failing to satisfy the Section 112(a) enablement requirement.  This trial resulted in a jury verdict that Idenix's '759 patent claims were not invalid for failure satisfy the enablement requirement of Section 112(a).  The Court granted Gilead's JMOL motion overturning the jury's verdict but denied Gilead's JMOL motion that the claims were invalid for failing to satisfy the written description requirement.  This appeal followed.

The Federal Circuit affirmed the District Court's JMOL decision on enablement, and reversed the District Court's denial of JMOL on written description, in an opinion by Chief Judge Prost joined by Judge Wallach; Judge Newman dissented.  The majority rendered its decision under the de novo review standard applied to JMOL motions, which permitted the appellate panel to more easily dismiss the jury's factual determinations.  The majority opinion characterized the issue before the Court as "whether a person of ordinary skill in the art would know, without undue experimentation, which 2'-methyl-up nucleosides would be effective for treating HCV."  The majority held that the answer to this question is no, because "a reasonable jury would not have had a legally sufficient basis to find otherwise."  The opinion rendered its decision by applying the factors delineated in In re Wands, 858 F.2d 731, 737 (Fed. Cir. 1988):

(1) the quantity of experimentation necessary;
(2) how routine any necessary experimentation is in the relevant field;
(3) whether the patent discloses specific working examples of the claimed invention;
(4) the amount of guidance presented in the patent;
(5) the nature and predictability of the field;
(6) the level of ordinary skill; and
(7) the scope of the claimed invention.

For context in appreciating how the majority applied the Wands factors, it is relevant to consider that the chemical arts have traditionally been considered unpredictable as compared with, for example, mechanical inventions.  While a mechanical device comprising a fastener, for example, could have as embodiments a handful of alternatives (a crew, a nail, a rivet, a bolt, glue, Velcro®), chemical compounds can have a multiplicity of substituents at a multiplicity of positions in a molecule, wherein the permutations can quickly exceed hundreds of thousands to millions, while but a few hundred exemplary compounds are disclosed in the specification.  The biotechnological arts are even more complex, for at least two reasons.  First, the molecules are even larger and have the capacity for additional substitutions, and the effects of those substitutions on function of a biological molecules are themselves unpredictable.  These scientific facts engendered the Federal Circuit's explication of the application of the written description requirement of Section 112 that culminated in the Court's en banc Ariad v. Eli Lilly decision (as well as earlier promulgation of Guidance from the U.S. Patent and Trademark Office in 2001).  Paradoxically, biotechnology patents (unlike chemical patents) do not disclose hundreds of exemplars (and frequently only one or a few), which has led to the scope of biotechnology claims to be relatively narrow.

These considerations provide an opportunity for the Federal Circuit to apply the factors set out in Wands stringently to find failure to satisfy the enablement requirement of Section 112(a), as the Court did here.  Going in order, the majority agreed with the District Court that the amount of experimentation required to support the "billions and billions" of putative species was high, supported by Gilead's expert testimony.  The District Court and the majority held that experimentation was too high even if mitigating circumstances would have presented a much smaller number of species (thousands) to the person of ordinary skill in the art.  This aspect of Idenix's argument was contradicted by its own evidence that "the field of modifying nucleosides for anti-HCV activity was 'in its infancy' and 'unpredictable'."  This conclusion was also supported by evidence that "many" of the candidate nucleosides would need to be synthesized because they were not commercially available, although the majority acknowledges that such synthesis was routine.

The majority then turned to the "working examples" and "amount of guidance" factors, which the opinion not surprisingly held supported non-enablement.  The opinion asserts in support of this conclusion that "Claim 1 requires more than just an identification of 2'-methyl-up: it requires identification of which 2'-methyl-up nucleosides will effectively treat HCV" and that "[w]ithout specific guidance on that point, the specification provides "only a starting point, a direction for further research," citing ALZA Corp. v. Andrx Pharm., LLC, 603 F.3d 935, 941 (Fed. Cir. 2010).  The (un)predictability prong of the factors was supported by trial testimony from both parties' experts, and the claim scope prong (essentially overbreadth) followed from the majority's conclusions regarding the rest of the factors.  The opinion's discussion characterized the situation as the person of skill in the art "the "large number" of 2'-methyl-up nucleosides falls into the 'small' group of candidates that effectively treats HCV."

As a consequence of these analyses, the majority readily concluded that the District Court correctly granted JMOL because no reasonable jury could conclude other than that Idenix did not satisfy the written description requirement.  The opinion notes that their decision has "striking similarities" to Wyeth and Cordis Corp. v. Abbott Laboratories based on the "millions of compounds made by varying the substituent groups" in that case wherein "only a 'significantly smaller' subset of those compounds would have the claimed 'functional effects'."  The opinion says that the decision here, as in Wyeth, "rests on the 'limits on permissible experimentation'," and states the somewhat new principle that "[w]here, as here, 'practicing the full scope of the claims would have required excessive experimentation, even if routine,' the patent is invalid for lack of enablement."

Turning to the written description issue, the majority readily pivoted from its enablement decision to hold that the '759 patent specification fails to provide an adequate written description because there was insufficient evidence that the Idenix inventors possessed the invention throughout its full scope.  In particular, the majority held that there was no evidence that the '759 inventors were in possession of Gilead's product.  As has been the case since the Federal Circuit's seminal decision in Regents of the University of California v. Eli Lilly, the absence of explicit disclosure of this species, in the further absence of a sufficient number of species to define a genus comprising Gilead's species, or structure/function relationships that would ensnare this species within the scope of the species expressly disclosed, was enough for the majority to conclude that the specification failed to satisfy the written description requirement.

The majority rejected Idenix's argument that the specification provided "abundant traditional blazemarks for the claims—working examples, formulas, data, synthesis routes, and the target," stating that the flaw in this analysis was that Idenix provided "lists or examples of supposedly effective nucleosides, but do not explain what makes them effective, or why."  In almost the reverse of the majority's reasoning regarding enablement, the opinion states that "the specification lists tens or hundreds of thousands of possible nucleosides, substituent-by-substituent, with dozens of distinct stereochemical structures, and yet the compound in question is conspicuously absent."

Judge Newman dissented; it must be said that the tone of the dissent, and that of a footnote in the majority opinion regarding the dissent, denotes a certain impatience on the part of both authors.  Judge Newman contends that "[t]he large number of unclaimed chemical variants in the specification are not described, not synthesized, and not tested for antiviral activity" and thus "[i]t is incorrect to include these variants in the claims and then to invalidate the claims because these variants are not described and not enabled."  The Judge believes that a reasonable jury could have considered the claims as being limited to the much smaller number of species exemplified in the specification and thus both enabled and adequately described.  She characterizes the majority's enablement theory as flawed for requiring description of "unclaimed and unsupported subject matter," and states that "a reasonable jury could have understood that subject matter that is unclaimed is irrelevant to validity under section 112."

In Judge Newman's view, the claims are limited by what is exemplified in the specification, wherein interpreting claim scope necessarily restricts the scope to that disclosure.  This is certainly a more parsimonious interpretation than the majority's and has the advantage that it would guard against a patentee expanding the scope of a claim to encompass species that a conscientious competitor pursues in an effort to avoid the claim.  The dissent recites copiously (18 separate citations, with the opinion stating there is "much more" ) from the expert testimony in this regard.  Judge Newman asserts that "[i]t was undisputed that the '759 specification did not describe and enable products other than those whose synthesis and antiviral properties were shown in the specification, all of which had the narrow formula of three OH groups and a CH₃ group as pictured.  A reasonable jury could have so viewed the claims."  She further states the jurisprudential principle that "[c]ourts are not free to reweigh the evidence and set aside the jury verdict merely because the jury could have drawn different inferences or conclusions or because judges feel that other results are more reasonable," citing Tennant v. Peoria & P.U. Ry. Co., 321 U.S. 29, 35 (1944).  Judge Newman concludes her dissent by stating that, despite Gilead's stipulation of infringement, the proper outcome of this case would be that the '759 patent claims were not invalid (when properly cabined to the scope supported by the specification) and not infringed by Gilead's fluorinated product (based on testimony as well as the absence of this species in the '759 disclosure).

In her own way, Judge Newman is putting her appellate thumb as heavily on the scale as did the majority, but in contrast, her jurisprudence would preserve the patent within the scope of the disclosure while absolving Gilead of infringement, while the majority's approach seems to be to interpret the claims broadly to reach the conclusion that they are invalid.  This decision continues the appearance, illustrated most starkly in the Court's decision denying rehearing en banc in Athena Diagnostics v. Mayo Collaborative Services, that the Court is seriously fractured in how it approaches its role as principle arbiter of U.S. patent law.

Idenix Pharmaceuticals LLC v. Gilead Sciences Inc. (Fed. Cir. 2019)
Panel: Chief Judge Prost and Circuit Judges Newman and Wallach
Opinion by Chief Judge Prost; dissenting opinion by Circuit Judge Newman

Comments

[Jim Demers]

Footnote 3: "The dissent, making an argument not advanced by Idenix at trial or before us, reaches the opposite conclusion only by disregarding the district court’s binding claim construction, ignoring the resulting stipulation of infringement, and analyzing a case that is not the one presented to us."

Kind of an important point, innit? The claim construction agreed to by the parties is bizarre: the claim plainly recites a "nucleoside", yet the construction throws away the 2" hydroxyl and brings 2'-deoxynucleosides within the scope. This may have suited their respective litigation strategies, but it's factually wrong, and it's what leads to the odd decision arrived at by the majority.

[Kevin E Noonan]

I supposed it depends on whether you think the court is there to come to the right decision or to decide the case before them warts and all. The only reason for the former at the Federal Circuit is the effect of a precedential decision on the rest of patent law. Certainly ways to decide this case without these consequences (making it non-precedential, for example).

[MaxDrei]

This opinion and the dissent remind me of those few well-reported cases in which a single patent issued by the EPO is litigated in various national jurisdictions, all within the EPC-space where the rules of claim construction are common to all national courts but where the courts differ in their opinions.

But I cannot recall any such cases in the very recent past, where a court in one EPC State decides "infringed but invalid" even while one in another EPC State decides "not invalid but not infringed either".

Perhaps that is because the judges in national patents courts in EPC-land really do not like disagreeing with each other in public, but very much enjoy arguing with each other in private, so regularly caucus together (like all academics) to debate, score points, and clarify their thinking.

One would think that a powerful dissent to a Federal Circuit opinion would help SCOTUS to bring the needed legal clarification. Instead, such a dissent seems only to confuse everybody.

[peter]

hmm