CPLRG™ 0016 - Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co. - Mar. 22, 2010

Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co., 598 F.3d 1336 (Fed. Cir. 2010) (En Banc) (LOURIE, Michel, Newman, Mayer, Bryson, Gajarsa, Dyk & Prost; NEWMAN, additional views; GAJARSA, concurring; RADER & Linn, dissenting-in-part & concurring-in-part; LINN & Rader, dissenting-in-part & concurring-in-part)

Ariad03222010

MAJOR ISSUES: Written description of invention (Section 112); requirement separate from enablement; original claims; support for generic claims; invention described by biological function

In this en banc decision, the Federal Circuit confirms existing law on the written description requirement but drew a sharp dissent from Judges Rader and Linn. The court offered some new guidance on when a specification disclosure can sufficiently support a generic claim. On written description, see Chisum on Patents 7.04; Chisum Patent Law Digest 2200. For an essay on Ariad (2010), see Chisum,Ariad(2010) and Written Description; Invention Priority Principles

1. REGULATING TRANSCRIPTION FACTOR. A patent concerned a method of reducing the activity of a transcription factor that regulates the extent to which cells express genes to produce certain proteins. U.S. Pat. No. 6,410,516.  U.S. Pat. No. 6410516

     a. NF-KB. The inventors discovered the factor, which they named “NF-KB.”

          i. FACTOR INDUCES GENE EXPRESSION. Once activated by a harmful extracellular influence, the NF-KB factor induces expression of numerous genes to produce proteins. The proteins help the cells survive but are harmful in excess.

          ii. BROAD CLAIMS: METHOD OF REDUCING NF-KB ACTIVITY. The patent’s claims were broad. They encompassed methods for “reducing binding of NF-KB to NF-KB recognition sites.”

          iii. SPECIFICATION: NO WORKING EXAMPLES; HYPOTHESES: THREE CLASSES OF MOLECULES COULD REDUCE NF-KB ACTIVITY. The patent’s specification disclosed no working examples. It hypothesized that three classes of molecules were “potentially capable of reducing NF-KB activity”: (1) specific inhibitors, (2) dominantly interfering molecules, and (3) decoy molecules.

     b. The patent owners, three academic institutions, and their exclusive licensee, sued an accused infringer, Eli Lilly, in a district court, alleging that the accused infringer’s two drug products, Evista® and Xigris®, infringed.

          i. ACCUSED INFRINGER: NO WRITTEN DESCRIPTION? JUST A RESEARCH PLAN? ROCHESTER (2004). The accused infringer argued that the patent’s asserted claims were invalid for, inter alia, lack of written description. It contended that the patent set forth “little more than a research plan, and does not satisfy the patentee’s quid pro quo as described in” University of Rochester v. G.D. Searle & Co., 358 F.3d 916 (Fed. Cir. 2004).

          ii. JURY VERDICT; APPEAL. A jury rendered a verdict for the patent owners, finding infringement and no invalidity. The accused infringer appealed.

2. HELD: the patent’s asserted claims are invalid for failure to meet the Section 112 written description requirement.

     a. SECTION 112, FIRST PARAGRAPH, CONTAINS A WRITTEN DESCRIPTION REQUIREMENT SEPARATE FROM ENABLEMENT.  The court “reaffirm[s] that § 112, first paragraph, contains a written description requirement separate from enablement ….”

     b. On whether original claims comply with the requirement, the court noted that many will but “certain claims may not.”

3. ARGUMENTS. In defense of its patent and the jury’s verdict upholding it, the patent owner posed alternative arguments, one legal and one factual.

     a. LEGAL ARGUMENT: FOR METHOD CLAIMS NOT REFERRING TO A COMPOUND, NO COMPOUND NEED BE DESCRIBED? The patent owner argued that, as a legal matter, because the asserted claims were to methods that recited no molecules, no description of molecules need be provided. The argument lacked merit because cases, such as University of Rochester v. G.D. Searle & Co., 358 F.3d 916 (Fed. Cir. 2004), had held “very similar method claims invalid for lack of written description.”  The patent owner argued that Rochester was distinguishable because the method claim in that case was to a method comprising a “administering a non-steroidal compound.” In contrast, the asserted claims in this case recited no compound or molecule.  The attempted distinction was unavailing. The specification “must describe some way of performing the claimed methods,” and it did so only through potential use of the three classes of molecules.

     b. FACTUAL ARGUMENT; SUBSTANTIAL EVIDENCE SUPPORTING JURY VERDICT. The patent owner argued that the specification, together with the testimony of its expert, Kadesh, “provided the jury with substantial evidence of adequate written description of the claimed methods.” The argument lacked merit.

          i. MUCH EVIDENCE IRRELEVANT BECAUSE AFTER THE FILING DATE. Much of the evidence the patent owner relied upon was “legally irrelevant” to whether the specification provided an adequate description to a person skilled in the art because it was dated after the patent’s effective filing date (April 21, 1989).

               A. DISPUTE OVER EFFECTIVE FILING DATE; JURY FINDING OF 1989, NOT 1991; NO APPEAL. At trial, the patent owner and accused infringer disputed whether the effective filing date was in the 1989 date or the date of a later application in 1991. In response to an interrogatory, the jury chose the 1989 date. Neither the patent owner nor the accused infringer appealed.

              B. Because the relevant date for written description was 1989, a figure setting forth a DNA sequence for a specific inhibitor (I-KB) added to the patent’s specification in the 1991 filing was not probative of written description. Also not probative was a 1990 publication on using decoy molecules to reduce NF-KM activity.

          ii. ADEQUACY OF DISCLOSURE OF THREE CLASSES OF MOLECULES TO SHOW POSSESSION OF THE METHOD. The specification’s description of the three classes of molecules, in light of the expert testimony, did not support the jury’s verdict that “the inventor possessed the claimed invention.”

               A. FIRST CLASS: “SPECIFIC INHIBITORS”; VAGUE FUNCTIONAL DESCRIPTION; FIGURE GIVING DNA SEQUENCE: TOO LATE AND INADEQUATE. A “specific inhibitor” is a molecule that can block NF-KB binding. The only example the specification gave was I-KB, a naturally occurring molecule. Most of the disclosure pertaining to I-KB was in the figure added in 1991 and, therefore, not relevant. Also, the figure was incomplete and inaccurate, a fact that led to a charge of inequitable conduct. See § 4261. The “vague functional description” of I-KB in the 1989 was only “an invitation for further research,” and, thus, not a sufficient written description of a specific inhibitor.

               B. SECOND CLASS: “DOMINANTLY INTERFERING MOLECULES”: OPERABLE ONLY IF FUTURE RESEARCH SHOWED THAT NF-KB DOMAINS WERE DISTINCT; ACTUAL PRACTICE SOON AFTER FILING. A “dominantly interfering” molecule is one truncated to retain a DNA binding domain but omit an RNA activating domain. The specification gave no example and noted that such a molecule would work only if the NF-KB domains were spatially distinct. Thus, the disclosure was merely a plan for further research and, therefore, not a sufficient description under Regents of the Univ. of Cal. v. Eli Lilly & Co., 119 F.3d 1559, 1566-67 (Fed. Cir. 1997). That workers actually practiced the technique soon after the patent’s 1989 filing date was not sufficient.

               C. “DECOY MOLECULES”; HYPOTHETICAL EXAMPLE; PROPHETIC EXAMPLE MAY BE SUFFICIENT BUT, HERE, THE SPECIFICATION MERELY MENTIONS A DESIRED OUTCOME. A “decoy” molecule is one having a gene region that mimics one that the NF-KB factor would normally bind to. Unlike the first two classes, the specification proposed a hypothetical example by disclosing DNA sequences. However, the specification merely stated that “NF-KB `would bind the decoy’ and thereby, `negative regulation can be effected.’ ” A “prophetic” example may satisfy the written description requirement, but, here, the specification merely mentioned “a desired outcome.” It did not provide a “descriptive link between the table of decoy molecules and reducing NF-KB activity.”

                D. VAST SCOPE OF GENERIC CLAIMS.  The court concluded that “[w]hatever thin thread of support a jury might find in the decoy-molecule hypothetical simply cannot bear the weight of the vast scope of these generic claims.

©2010 Donald S. Chisum - All Rights Reserved

Website design by Bluegrass Internet Services