Docket No. 2012-1433
RADER(d), SCHALL, BRYSON
July 22, 2013
Brief summary: Novozymes’ general description of mutations (any mutation at residue 239) was insufficient to show possession of invention claimed in continuation (filed “[u]pon learning” of DuPont’s mutant).
Summary: Novozymes appealed DC grant of DuPont’s post-trial motion for JMOL that the claims of Novozymes’ US 7,713,723 relating to alpha-amylase enzyme preparations for breaking down polysaccharides (e.g., for making detergent formulations, refining sugars, ethanol production, etc.) are invalid under § 112 (written description). Novozymes markets a product derived from Bacillus licheniformis (BLA) as TermamylTM. Novozymes’ provisional application (“2000 application”, identical to the ‘723 specification) describes BLAs with enhanced stability resulting from modification of amino acid sequences in the proteins. The application described seven parent enzymes, identified thirty-three positions as “promising mutation targets” which could be modified by deletion, addition or substitution to exhibit improved stability at low calcium concentrations. Sixteen positions for possible mutation were identified by rational design and 17 by random mutagenesis. None of the positions or combinations of mutations were identified as preferred. Data showing improved stability from the 17 positions identified by random mutagenesis but none was presented for those identified by rational design. During Novozymes’ prosecution of its application, DuPont began working on mutants, identified the S239Q mutant and obtained US Patent No. 7,541,026. Position 239 was identified in Novozymes application but the specific S to Q substitution was not. Novozymes filed a continuation application directed to mutants at residue 239 in 2009 “[u]pon learning” of DuPont’s mutant, which ultimately issued as the ‘723 patent. Disputed ‘723 claim 1 claimed “[a]n isolated variant” having “at least 90% sequence identity to [parental] SEQ ID NO:6”, “compris[ing] a substitution of serine at position 239”, and having “increased thermostability relative to the parent alpha-amylase”. The opinion points out that “[e]ach of those limitations can be found at points within the underlying 2000 application, but, outside of the ‘723 patent’s claims, Novozymes never presented them together in any particular embodiment and did not highlight the BSG parent or position 239 among the other disclosed options.” At the DC, the jury found DuPont liable for willful infringement but DuPont was granted JMOL for lack of written description. In its decision, the DC “noted that the 2000 application disclosed a potentially enormous number of alpha-amylase variants” and “did not point out the specific variants later claimed in the ‘723 patent”, citing Boston Scientific (FC 2011), Centocor (FC 2011), Univ. Rochester (FC 2004), Purdue (FC 2000) and In re Ruschig (CCPA 1967). In the appeal, Novozymes argued that the 2000 application expressly disclosed each limitation of the asserted claims and that the technology was not unpredictable (citing Union Oil (FC 2000)). The FC panel disagreed, however, holding that “the supporting disclosure of the 2000 application provides only generalized guidance listing several variables that might, in some combination, lead to a useful result”, “nothing in the 2000 application indicates that Novozymes then possessed what it now claims”, and “the testimony of Novozyme’s experts does not overcome the fundamental deficiencies of the 2000 application’s written description”, even though “the 2000 application provides formal textual supported for each individual limitation recited in the claims”, citing In re Ruschig (“a claim to a specific drug molecule, added after filing” not supported by “only a generic structure that could yield the claimed molecule”), Boston Scientific (no WD with “passing reference to the term ‘macrocyclic triene'” and failure to “identify any member of the claimed sub-genus” or analogs), Purdue Pharma (no WD where two of seven examples “could be shown to meet the claimed ratio limitation by piecing together the disclosed data” but did not “in any way emphasize the [claimed] ratio”), and University of Rochester (no WD for drugs “having a certain, selective activity” but no disclosure of any such drugs and none known in the art). It also reviewed cases with opposite outcomes including Snitzer (fourteen types of lasers disclosed, several inoperative, but “literal description of the ytterbium ion provided adequate support for claiming that species”), In re Driscoll (WD where “the particular claimed compound had been individually described as one of several possibilities”), and Union Oil (WD where “ordinarily skilled petroleum refiners would immediately appreciate that the quantitative chemical properties recited in the claims translated to specific, manifest compositions that would yield those properties”). The opinion also stated that “the ‘723 patent lacked meaningful support”, the application did not “verif[y] whether any of the remaining seventeen positions predicted by rational protein design (including position 239) actually yielded a thermostable variant”, and the skilled artisan would only have understood that the application “predicted that at least some mutations at position 239 would yield variants with increased thermostability, not that it had possessed or had definitively identified any mutations that would do so.” And, “[i]n this case, to actually possess the variant enzymes…would have required Novozymes to confirm its predictions by actually making and testing individual variants or at least indentifying subclasses of variants that could be expected to possess the claimed properties….” Judge Rader’s dissent argued that the jury’s factual finding of satisfactory written description “deserves significant deference” and that the DC decision should have been reversed.