For a broad reference to be considered as disclosing a particular chemical compound for prior art purposes, it must “teach a finite and limited class” that includes the compound claimed. Here, for example, a compound representing the active ingredient in a claimed pharmaceutical was found to not be disclosed by a formula covering compounds from a first list of 17 amino acids, a second list with over a hundred possible combinations of amino acids and protecting groups, and a third list that was open-ended. “This too is not a definite and limited class.” This would be a good case to consult and cite in response to an anticipation or obviousness rejection predicated on an expansive listing of materials that may happen to overlap the composition claimed.
Background / Facts: The patents being asserted here in response to a generic drug ANDA submission are directed to derivatives of amphetamine that are modified to decrease their activity when administered in high doses—as happens when the drug is being abused. The active pharmaceutical ingredient is L-lysine-d-amphetamine (“LDX”) dimesylate. The primary reference in an assertion of invalidity discloses combining amphetamine, in any of its stereochemical forms, with numerous amino acids, in various stereochemistries and with many potential protecting groups.
Issue(s): Whether such a broad disclosure is sufficient to disclose or even suggest the active ingredient LDX as claimed.
Holding(s): No. “[The primary reference] lists 18 amino acids ‘and the like,’ and states they can belong to the D- or L-series. Even this list, therefore, does not limit itself to 18 amino acids. [The primary reference] expressly suggests post-translational modifications of the amino acids [], thus further increasing the potential amino acid groups to be utilized. While [the cited portion] states that ‘[a]cids of the L-series are preferred,’ [it] actually describes numerous D-series amino acids. Read in context of the whole reference, a person of skill in the art would, therefore, not focus exclusively on amino acids with the L stereochemistry.” Further, the primary reference “suggests that [its Chemical Formula IV] can be selected from one of three lists, and as defendants’ expert candidly admitted, Formula IV ‘does not indicate any preference’ among the different options. Thus, Formula IV discloses all the compounds from all three lists, the first of which lists 17 amino acids (including lysine), the second of which teaches over a hundred possible combinations of amino acids and protecting groups and the third of which does not even provide a definite list of compounds. This too is not a definite and limited class.”