Earlier this month, the Federal Circuit heard oral argument in Alnylam Pharmaceuticals, Inc. v. Moderna, Inc., a patent case we have been following because it attracted an amicus brief. In it, Alnylam Pharmaceuticals, Inc. appeals a judgment of a district court based on the argument that the court made an error in claim construction. Judges Taranto, Chen, and Hughes heard the oral argument. This is our argument recap.
Paul W. Hughes appeared on behalf of Alnylam Pharmaceuticals. He opened by arguing that “this is a claim construction dispute and, as always, the court should start with the language of the claims.” He explained how, “here, the claim language read with the whole specification makes abundantly clear that the heartland of these claims encompasses branched alkyls with two carbon bonding, because the branching occurs at the alpha position.” In response, a judge asked, “if that’s true, what is that definition doing in the specification?” Hughes responded by arguing that the specification is “talking about the underlying science” and “functions as a textbook.” He explained that, “when you look at the patent . . . the overwhelming majority of examples describe a branched alkyl where the branching occurs within the alkyl chain.” And, he argued, “the only way that you can have a branched alkyl, when you have two carbons, is where the branching occurs in the alpha position, because . . . it’s not in the chain.”
Jeffrey A. Lamken appeared on behalf of Moderna. He began by arguing “the patent’s definition of branched alkyl is unmistakable lexicography.” He explained that the definition “not only appears in the dictionary section; it not only says branched alkyl and sets it off in quotation marks;” but it “uses definitional phrasing–it says ‘refers to,’ which is typically something that means it is a definition.” He explained that the language “provides a clear definition with two components. First, he said, “you have to have one carbon bonded to three others.” Second, he continued, you must have “an exclusion as well.” He explained that “the qualifier, ‘unless otherwise specified,’ tells the skilled artisan how to recognize any exceptions.” Ultimately, he argued, the exceptions “will be specified, which means it’s specific, clear, unambiguous.”
Lamken suggested “the patent draftsman knew how to make something non definitional, but instead” used the language “refers to” to create a definition that “serves a critical notice function.” He argued a “skilled artisan is entitled to look at that definition” with the understanding that it applies.” He conceded that usually with this type of claim one sees a structure or diagram, but here, he highlighted, “they chose to describe it using words.” He argued “if you’re going to use words in the claim to describe a chemical structure, you’re going to have to define the words and give them meaning.” According to Lamken, “the specification here goes and gives us a definition that’s absolutely clear beyond doubt” that to infringe one needs “three carbons.”
In his rebuttal, Hughes argued Moderna’s position is that the only work claim twenty can do is a hypothetical triple bonding. But, he argued, the specification provides “no indication of triple branching at the alpha position.” He indicated “the patentee made clear throughout” the specification that the relevant formula “can only be read as showing that you can have two carbon bonding at the alpha position.” He suggested that, “ultimately, the claim language is what matters,” and, he continued, the “claim language here has two important features.” First, he explained, “the claim language has its branching at the alpha position.” Second, he continued, “every piece of the specification makes clear that one skilled in the art would know that two carbon bonding is appropriate.”
We will continue monitoring this case and report on developments.