A three-judge panel of the Court of Appeals for the Federal Circuit recently decided Ariad Pharmaceuticals, Inc. et al. v. Eli Lilly & Company, appeal no. 2008-1248 (Fed. Cir. April 3, 2009), involving Blatimore et al. (U.S. Pat. No. 6,410,516). The “et al.” is M.I.T., The Whitehead Institute and Harvard University. Despite the impeccable scientific credentials of the inventors, that included two Nobel Prize winners, the panel reversed the district court’s denial of Lilly’s motion for JNOV of a jury verdict finding of validity and infringement of the claims in suit. A representative “composite” claim (80) reads as follows:

A method for modifying effects of external influences on a eukaryotic cell, which external influences induce NF-kB-mediated intracellular signaling, the method comprising reducing NF-kB activity in the cells by reducing binding of NF-kB to NF-kB recognition sites on genes which are transcriptionally regulated by NF-kB, such that NF-kB-mediated effects on external influences are modified.

Claims 144-145 are similar but recite a more specific effect – that the method reduces bacterial LPS induced expression of cytokines in mammalian cells. Since Patents4Life began a few days before this decision, the reader is respectfully requested to refer to the helpful commentary in Nature/Biotechnology, 24, 737 (Jul 2006), in which I am extensively quoted.

The jury found that the drug Evista infringed claims 80 and 95 and that the drug Xigris infringed claims 144 and 145. The plaintiff’s victory was in jeopardy from the start, as the panel recited the principles of law controlling the written description requirement (WDR) of 35 U.S.C. 112 for three pages of the slip opinion. Lilly had argued that the claims are not supported by an adequate written description of the method, since the patent does not teach how to reduce the activity of NF-kB. The specification only disclosed three functionally-named classes of inhibitor compounds, e.g., “decoy molecules”. Ariad argued that the claims could not fail the WDR since they did not recite any inhibitor compounds (in the claims), and the patent did not purport to claim any such molecules.

The Federal Circuit made short work of Ariad’s argument, essentially finding less description than they found to be inadequate in the U. of Rochester decision (the “COX-2” patents) in which the claims-in-suit recited a “non-steroidal compound” possessing certain activity, but provided no examples of such compounds.

Ariad also advanced a procedural argument – that they had presented the jury with “substantial evidence” of adequate description via an expert’s interpretation of the specification at trial. The panel found that much of the evidence was directed to the state-of-the art as it was after the 1989 date chosen by the jury as the effective filing date of the patent. The panel went on to carefully examine the specification, and found a lack of substantial evidence to support the verdict that the patent’s written description showed possession of the inhibitors by the inventors. The panel noted that the only inhibitor named was the “I-kB” molecule that binds to NF-kB to hold it in an inactive state in vivo until signals are received to release it. The panel found that the I-kB sequence was not disclosed in the 1989 application and, adding insult to injury, found that the inventors had inadequately described the sequence when they added it in a figure (43) in a later filing.

The panel reviewed the Federal Circuit’s recent (post-UC v. Lily) WDR jurisprudence, seemingly with relish when they wrote that the putative disclosure of one class of inhibitor in the patent “just represents a wish or arguably, a plan for future research” to obtain it. The panel noted that the specification contained examples of DNA sequences that were disclosed to be useful as “decoy molecules” but was unconvinced that the specification adequately taught how to use them. The panel found that the specification did not even disclose a “hypothetical example” of how to use them. The panel’s comment that hypothetical examples “certainly can be sufficient to satisfy the written description requirement,” could be a backhanded attempt to shore up the Kubin decision, in which the primary prior art, the Valiente patent, contained a hypothetical example of how to isolate the protein that Kubin et al. actually characterized. In any case, the ‘516 patent had no such examples, and the panel found “gaping holes” in its disclosure that rendered it insufficient to support “the vast scope of these generic claims.”

The panel upheld the district court’s finding on no inequitable conduct on the basis that there was no clear and convincing evidence of intent to deceive by the attorneys who handled the application. The attorneys had deleted erroneous figure 43 in a number of applications, and their failure to do so in the application that issued as the ‘516 patent did not rise to the level of “purposeful concealment.” This was the case even though materiality was considered to be high. The court had also dismissed Lilly’s attempt to demonstrate inequitable conduct due to the failure of one inventor to disclose his later work on NF-kB inhibitors to the PTO. The court felt that the inventor probably did not understand the relevance of the work to an inherent anticipation theory that Lilly had advanced. The panel restated the standard that some amount of intent must be established by clear and convincing evidence, no matter how great the materiality of the omitted information [citing Star Scientific v. R. J. Reynolds, 537 F.3d 1537 (Fed. Cir. 2008) repeatedly].

In his concurrence, Judge Linn reiterated his view as set forth in Rochester, that establishing a separate WDR in 112 was unnecessary, when the enablement requirement was adequate to test patents such as the ‘516 patent. Judge Linn also urged the Federal Circuit to address the issue of whether a specification can enable “unknown methods … an important issue that we have left unresolved.” Indeed it is. This cogent concurrence will be a subject of future posts, as will be reexamination procedure that Merck initiated to advance the inherent anticipation arguments the panel also did not consider