Structure or Function? AbbVie Deutschland GmbH & Co. v. Janssen Biotech, Inc. and the Federal Circuit’s Structure-Function Analysis of Functionally Defined Genus Claims Under Section 112’s Written Description Requirement

I. Introduction

On July 1, 2014, in AbbVie Deutschland GmbH & Co. v. Janssen Biotech, Inc.,[ref]759 F.3d 1285 (Fed. Cir. 2014).[/ref] the Court of Appeals for the Federal Circuit affirmed a district court’s decision invalidating the claims of two antibody patents for failing to meet Section 112’s written description requirement.[ref]See id. at 1290.[/ref] Specifically, the court took issue with the patents’ functionally defined antibody-genus claims, concluding that the patentee failed to disclose a representative species of antibodies diverse enough to support the patents’ broad genus claims. Also of note was the court’s consideration of the preclusive effect of decisions by the Patent and Trademark Office’s Board of Patent Appeals and Interferences (Board). Moving forward, the AbbVie decision casts doubt on the validity of functionally defined genus claims, particularly in highly technical fields such as biological sciences, and it closes the door on parties attempting to make collateral-estoppel arguments based on interference decisions that are in the process of being challenged under 35 U.S.C. § 146.

II. Technology At Issue

The patents at issue in AbbVie, U.S. Patent Numbers 6,914,128 (‘128 patent) and 7,504,485 (‘485 patent), were directed to a class of fully humanized antibodies designed to bind to and neutralize Interlukin-12 (IL-12), a signaling protein secreted by the human body that, if overproduced, can lead to psoriasis and rheumatoid arthritis.

The structure of the claimed antibody genus played a major role in the court’s written description analysis. Generally speaking, an antibody consists of four chains of amino acids, two identical heavy chains and two identical light chains, folded into a three-dimensional structure that binds to an antigen and facilitates the removal of the antigen from the body. The heavy and light chains have both a constant and variable region. Among human antibodies, the variable region of the heavy chain is divided into seven families, ranging from VH1 to VH7. Similarly, the variable region of the light chain is divided into two classes, either Kappa or Lambda. The variable regions of the heavy and light chains each have three complimentary determining regions (CDRs). The binding affinity of an antibody—that is, its ability to bind to and later disassociate from the antigen—is measured by what is referred to as the antibody’s k_off rate. The lower the k_off rate, the tighter the antibody binds with the targeted antigen.

AbbVie’s ‘128 and ‘485 patents claimed the entire genus of fully humanized antibodies that bind IL-12, but it did so only by the antibodies’ functional characteristics. Said differently, AbbVie claimed the binding and neutralizing characteristics of the antibody, not the structural composition of the antibody itself. Claim 29 of the ‘128 patent, which was treated as representative, claimed the following: “A neutralizing isolated human antibody, or antigen-binding portion thereof that binds to human IL-12 and disassociates from human IL-12 with a k_off rate constant of 1×10^-2 s^-1 or less, as determined by surface plasmon resonance.”[ref]Id. at 1292.[/ref]

In its disclosure, AbbVie listed more than 300 different variations of antibodies that achieved the desired k_off rate. That said, all 300 of the disclosed antibodies variations were derived from the antibody Joe-9 and thus all shared the same structural makeup: VH3 heavy chains and Lambda Light chains. The differences between each of the 300 disclosed antibodies were only minor changes to the CDR sequences of the Joe-9-based antibodies.[ref]See id. at 1291.[/ref] Notably, although AbbVie disclosed these 300 variations, AbbVie chose not to disclose any structural features or characteristics that were common to the members of the claimed antibody genus.

The Defendant-Appellee, Centocor, markets its own antibody that neutralizes human IL-12 under the brand name Stelara®. Although the Stelara® product is comprised of VH5 heavy chains and Kappa light chains—an entirely different structure than the Joe-9-based antibodies disclosed in the ‘128 and ‘485 patents—AbbVie nonetheless alleged that Stelara® fell within the scope of the genus claims because it was a fully humanized antibody that bound to and disassociated from IL-12 at the claimed k_off rate.[ref]AbbVie Deutschland GmbH & Co., 759 F.3d at 1292 (outlining structural characteristics of Stelara®).[/ref]

III. Background

After filing its patent application for Stelara®, Centocor provoked an interference with AbbVie’s ‘128 patent to contest both priority and validity of the ‘128 patent on obviousness grounds. The Board found that the ‘128 patent was not invalid for obviousness and awarded priority to AbbVie. Shortly thereafter, AbbVie and Centocor each filed actions—AbbVie an infringement action in the U.S. District Court for the District of Massachusetts and Centocor a declaratory-judgment action for noninfringement and invalidity in the U.S. District Court for the District of Columbia. Separately, Centocor also challenged the Board’s earlier invalidity and priority decisions under 35 U.S.C. § 146, which allows a party to contest a Board ruling by filing a civil action in federal court. Centocor’s actions were transferred to the District of Massachusetts and consolidated with AbbVie’s infringement action.[ref]AbbVie Deutschland GmbH & Co. v. Janssen Biotech, Inc., 759 F.3d 1285, 1292-93 (Fed. Cir. 2014) (providing procedural history).[/ref]

At trial, AbbVie moved for summary judgment on the grounds that Centocor should be collaterally estopped from challenging the validity of the ‘128 patent because Centocor had failed to invalidate the patent in the parties’ earlier interference proceeding. The trial court denied AbbVie’s summary-judgment motion but, after construing the claims, found that Centocor nevertheless infringed the asserted claims of the ‘128 and ‘485 patents. With infringement decided, the issue of validity was tried before a jury.

Upon conclusion of the trial on validity, the jury determined that each of the asserted claims were invalid for lack of adequate written description, lack of enablement, and obviousness. AbbVie moved for a judgment as a matter of law (JMOL) on each of the invalidity grounds, but the motion was denied.[ref]Id. at 1290 (outlining issues at the trial court).[/ref] On appeal, AbbVie challenged the district court’s denial of summary judgment and JMOL.[ref]Id. AbbVie also challenged the district court’s denial of its motion for a new trial on the basis of evidentiary and jury-instruction errors. AbbVie did not appeal the obviousness determination outright but instead alleged that the jury-instruction error warranted a new trial. Id.[/ref]

IV. Collateral Estoppel

The collateral-estoppel argument AbbVie presented to the Federal Circuit focused on the parties’ earlier interference proceeding. AbbVie contended that the Board’s interference decision was final and thus Centocor should have been foreclosed from relitigating the issue of patent validity at trial. AbbVie believed that the Board’s validity ruling in the interference action was binding and Centocor’s Section 146 proceeding was akin to an appeal of an otherwise-final judgment. In response, Centocor argued that, unlike other types of appeals, Section 146 allows the party challenging a Board decision to put forth new evidence and have the entire record reviewed de novo by the district court. And thus, as long as the Section 146 proceeding was pending, Centocor contended, the Board’s decision was not yet a “binding final judgment” for purposes of collateral estoppel. The Federal Circuit agreed with Centocor, holding that because the district court can conduct a de novo review of all of the evidence in a Section 146 proceeding, including new evidence that was inadmissible at the Patent and Trademark Office (PTO), a Board decision cannot be considered a final judgment that would preclude a party from relitigating the same or similar issues in a parallel action.[ref]Id. at 1296-97 (noting Board decision is not a binding final judgment because “the factual record remains open with respect to the issues contested at the PTO . . . . [And] [b]ecause a district court can make a de novo determination of facts upon the submission of new evidence.”)[/ref]

Although the Federal Circuit made clear that a Board’s decision is not final when challenged under Section 146 for purposes of collateral estoppel, the court expressly stopped short of applying that same rationale to Board decisions challenged under 35 U.S.C. § 141.[ref]Id.at 1297 (“Whether a Board’s interference decision that is on appeal under § 141 can have collateral estoppel effect on issues raised in a co-pending litigation is another question, one we need not address here.”)[/ref] That section allows a party to appeal a Board’s ruling directly to the Federal Circuit. Moving forward, it will be interesting to see if the Federal Circuit treats Section 141 challenges to interference proceedings like typical appeals, allowing parties to argue that underlying Board decisions have collateral-estoppel effect in subsequent or parallel litigations.

V. Written Description

The primary focus of the Federal Circuit’s decision in AbbVie was its application of the written description requirement in 35 U.S.C. § 112 to AbbVie’s expansive antibody-genus claims. AbbVie contended that the district court should have granted its JMOL motion on the issue of written description. Specifically, AbbVie believed there was adequate support for its broad genus claim because AbbVie had disclosed more than 300 antibodies covering the full range of the claimed k_off rates in the patents’ specifications. And Centocor’s attempts to distinguish Stelara® on the basis of structure, AbbVie contended, were legally irrelevant because AbbVie’s claims were defined by k_off rate, not structural characteristics. In response, Centocor pointed out that, although AbbVie disclosed over 300 variations of antibodies, all of those variations were Joe-9-based; thus, all of the disclosed antibodies had VH3 heavy chains and Lambda light chains. Only small adjustments to the CDR sequence differentiated each disclosed antibody from the next. In sum, the thrust of Centocor’s argument was that AbbVie did not disclose, and thus could not prove it “possessed” the idea of, a sufficiently representative number of antibody species required to support such a broad genus of antibodies

The Federal Circuit has held that when a patent claims a functionally defined genus, “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.”[ref]Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1349 (Fed. Cir. 2010)[/ref] More specifically, to satisfy Section 112’s written description requirement, a party claiming an entire genus must disclose either: a representative number of species falling within the scope of the genus or structural features so common to the members of the genus so that one of skill in the art can visualize or recognize the members of that genus.[ref]Id. at 1350 (citing Regents of the Univ. of Cal. v. Eli Lilly & Co., 119 F.3d 1559, 1568-69 (Fed. Cir. 1997)).[/ref] On appeal, AbbVie conceded that it did not disclose structural features common to the members of the genus, arguing instead that its roughly 300 disclosed antibodies constituted a representative number of species to support its genus claims.

Despite AbbVie’s contentions that 300 disclosed antibodies were sufficiently representative, the Federal Circuit affirmed the district court’s denial of AbbVie’s motion for JMOL, holding that substantial evidence supported the jury’s verdict of invalidity for lack of adequate written description. The court explained that “the jury heard ample evidence that AbbVie’s patents only describe one type of structurally similar anti-bodies and that those antibodies are not representative of the full variety or scope of the genus.”[ref]759 F.3d at 1300.[/ref]

To illustrate its species-genus rationale, the court analogized a claimed genus with a plot of land, maintaining that “if the disclosed species only abide in a corner of the genus, one has not described the genus sufficiently to show that the inventor invented, or had possession of, the genus. He only described a portion of it. That is the case here.”[ref]Id.[/ref] Further, the court used this same analogy to address AbbVie’s argument that the company did not need to disclose structural characteristics of the species because its claim was functionally based.

One describes a plot of land by its furthest coordinates, in effect drawing a perimeter fence around it. That may be akin to the function of patent claims to particularly point out and distinctly circumscribe the outer boundaries of a claimed invention. With the written description of a genus, however, merely drawing a fence around a perceived genus is not a description of the genus.[ref]Id.[/ref]

In sum, the court held that although AbbVie did not have to disclose every possible species, or even the allegedly infringing compound itself, in exact terms to comply with Section 112’s written description requirement, “the patents [have to] at least describe some species representative of antibodies that are structurally similar to” the allegedly infringing compound.[ref]Id. at 1301.[/ref]

VI. Implications of AbbVie

Moving forward, the AbbVie decision provides clarity in two areas of the patent law. First, the decision makes clear that Board decisions that are in the process of being challenged under Section 146 will not be considered final binding judgments for collateral-estoppel purposes. That said, the court expressly side-stepped the issue of whether Board decisions that are challenged by means of Section 141—a direct appeal of the decision to the Federal Circuit—could have preclusive effect. Based on the court’s emphasis on the ongoing, open nature of Section 146 proceedings in reaching its conclusion, however, my expectation is that the Federal Circuit, when presented with the opportunity, will hold that Board decisions challenged under Section 141 do in fact constitute final binding judgments. Thus, such decisions could serve to preclude a party from later challenging those same issues in a subsequent or parallel proceeding.

Second, the decision raises the bar on patentees seeking to claim functionally. In the wake of AbbVie, patentees with functional claims will now need to go out of their way to establish “a reasonable structure-function correlation” in their specification or by reference to the knowledge of one skilled in the relevant art. And perhaps more notably, for existing patents, the decision unambiguously calls into question the validity of functionally defined genus claims. In fact, the court went so far as to say that functional claims are “inherently vulnerable to invalidity challenge[s] for lack of written description support, especially in technology fields that are highly unpredictable,” such as biological sciences.[ref]AbbVie Deutschland GmbH & Co., 759 F.3d at 1301. The court noted functional genus claims in technology fields will be particularly susceptible to validity challenges under Section 112’s written description requirement because “it is difficult to establish a correlation between structure and function for the whole genus or to predict what would be covered by the functionally claimed genus.”Id. (citing Ariad Pharm., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1351 (Fed. Cir. 2010). In Ariad Pharmaceuticals, the Federal Circuit wrote, “[T]he level of detail required to satisfy the written description requirement varies depending on the nature and scope of the claims and on the complexity and predictability of the relevant technology.” 598 F.3d at 1351.[/ref]

In light of the court’s express skepticism regarding functionally defined genus claims, my expectation is that challenges to the validity of these types of patents on written description grounds will continue to grow. And, generally speaking, I think that is a good thing. Allowing patentees to claim a broad genus without appropriately disclosing species sufficient to support those claims almost certainly gives rise to undeserved patent protection on inventions that simply were never invented. Indeed, such a grant is at odds with the very purpose of the patent laws—a patentee should not be granted a monopoly on an invention he or she has yet to invent or fails to appropriately disclose to the public. I also believe the decision could be a boon for pharmaceutical innovation. As these broad patents begin to fall, I expect that pharmaceutical manufacturers will have a greater incentive to innovate in various areas of biological science that until now fell within the scope of broad, functionally defined genus claims of various biological-science patents.