Sublingual Dosage Form Patent Valid but Not Infringed on the Facts

Meda AB v Pharmascience Inc 2016 FC 1362 Manson J
            2,629,988 / zolpidem / SUBLINOX / NOC

In this NOC proceeding, Meda’s ‘988 patent was held to be substantially valid, but not infringed by Pharmascience’s PMS-Zolpidem product. This result turned entirely on the facts, and no significant points of law were raised. The overbreadth discussion does provide an interesting example of the role of claim differentiation.

The ‘988 patent is a formulation patent, claiming zolpidem for the treatment of insomnia in a sublingual dosage form comprising an ordered mixture, in which the API is adhered to larger carrier particles, and a bio/mucoadhesive promoting agent [76]. (As I understand it, an “ordered mixture” means a homogeneous mixture, as opposed to a random mixture in which some degree of segregation may occur.

Validity was attacked on the basis of anticipation, obviousness, lack of utility and overbreadth [18]. The anticipation argument failed because the prior art relied on did not teach an ordered mixture or sublingual administration [134]. The patent is a combination patent, in the sense that all the elements were known in the prior art [77], and the question was whether it would have been obvious to combine them. On the facts, Manson J held that the combination was not obvious [154]. Pharmascience’s argument to the contrary was not helped by the fact that in a review article its own expert had described the ‘988 patent as “an unusual approach” and “very different from all the formulations that [he’d] read” [151]. The utility attack was based on the promise of bio/mucoadhesion. Manson J held that “[a]ll that is required is sufficient bio/mucoadhesion to achieve the promised rapid release and absorption” [176] and this promise was demonstrated by examples in the patent [178].

Pharmascience also attacked the patent on the basis of overbreadth. Specifically, Pharmascience argued that the claims were overbroad because it was necessary that the formulation (1) have the bio/mucoadhesive agent on the surface of carrier particles; and (2) be essentially free from water [157], and Claim 1 was invalid because it did not have the first limitation, and all the claims were invalid because none of them had the second limitation. Both limitations were, however, disclosed in the specification, and Meda responded that a POSITA reading the claims in light of the specification would understand those limitations to be implicit [162]. Manson J agreed that a skilled person reading the claim “in light of the specification as a whole, with a mind willing to understand, would know that the formulation would have to be essentially free from water to work” [168]. The attack based on the absence of that limitation therefore failed. However, Manson J held that it was not permissible to read limitation (1) as being implicit in Claim 1, because it was explicit in Claim 2 and following claims, and “the presumption against redundancy precludes this limitation from being implied into claim 1" [168]. Consequently, Claim 1 was invalid for overbreadth [168]. However, this did not substantially affect the strength of the patent as a whole, because Claim 2 was exactly the same as Claim 1 except that it did include this limitation. The interesting point is what would have happened if Claim 2 had omitted this limitation; it seems not unlikely that Manson J would have been willing to find it to implicit, as he did with the requirement of being free from water.

While the patent was held to be substantially valid, Manson J held that Pharmascience’s allegation of non-infringement was justified [195]. The reasoning on this point is not entirely clear, as the discussion was substantially redacted. It is nonetheless clear enough that the result turned on the facts.