The Limits of Sufficiency

Allergan Inc. v. Juno Pharmaceuticals Corp 2023 FC 1686 Pentney J

2,585,691 / bimatoprost formulation / LUMIGAN RC / NOC

Glaucoma and ocular hypertension are serious medical conditions that can lead to blindness. It can be controlled by medications which lower the pressure inside the eye, referred to as intraocular pressure (“IOP”) [4]. “[T]aking glaucoma medication is boring, because it requires a daily dose of one or more medications, often for the rest of a person’s life. And the patient will usually not notice any improvement in their vision, because the medication is actually only stopping further decline” [48]. For this reason, patient compliance is very important. (I must say that I enjoyed Pentney J’s lucid discussion of the physiology of the eye and drug transport through it [29]–[49]. One of the reasons I specialize in patent law is that the science is interesting and Pentney J’s discussion was particularly clear and informative.)

Allergan’s old LUMIGAN product was used to treat IOP. It contained 300 ppm bimatoprost, the active ingredient, and 50 ppm BAK, a preservative. It was successful in lowering IOP but had unwanted side effects, which impacted compliance. “Allergan therefore sought to develop a formulation that was similarly effective in lowering IOP but did not cause the unwanted side effects” [17]. The result was the claimed invention, embodied in LUMIGAN RC, which contained 100 ppm bimatoprost and 200 ppm BAK—that is, it had less of the active ingredient and more preservative.

The main question was whether this formulation was obvious. Pentney J helds on the facts that it was not. While reducing the concentration of bimatoprost might be expected to reduce side effects, the skilled person would also expect it to reduce efficacy [203]. While BAK is generally safe at the levels in both the old and new formulations, there was nonetheless a trend to decreasing BAK levels, to minimize its potentially inflammatory and cytotoxic effects [241], [243]. The key to the formulation is that BAK also worked as a “penetration enhancer” which meant that the lower amount of bimatoprost in the claimed formulation achieved equal or better IOP reduction than old formulation [245]. This effect was not obvious. It was known that BAK could be a penetration enhancer for some molecules, in particular hydrophilic molecules, but it was not expected to be a penetration enhancer for lipophilic substances such as bimatoprost [302]–[305]; and see [319]. Consequently, Pentney J found the claimed formulation was not obvious after an obvious-to-try analysis [380]–[382].

There was a fairly extended discussion of the inventive concept and whether it should be determined from the claims alone or by considering the disclosure as well. Pentney J rejected an approach based on the claims alone, noting that Shire 2021 FCA 52 [72] held that it is necessary to turn to the specification, at least in the case of a claim to a chemical compound [158]. Pentney J noted that Shire was particularly pertinent to the case at hand [153], because a chemical formulation is no different from a chemical compound in that respect. Consequently, Pentney J embarked of a fairly lengthy examination of the disclosure in light of the expert evidence in order to identify the inventive concept.

I won’t go through his discussion in detail, but I will point out that looking at the claims and looking at the specification are not the only two possible choices. The problem with looking at the claims is that a claim to a compound does not tell us what the inventive step was: the compound might be easy to synthesis, but with no previously known use, in which case identifying the use might be inventive. Or a compound might be known to be useful, but difficult to synthesize (eg early biologics), in which case the method of synthesis might be the inventive step. We cannot tell which by looking at the claims alone.

The problem with looking at the disclosure to determine the inventive concept is that the inventor does not always know what they have invented, because the inventor does not normally know all the prior art, and may well not know the entire state of the art (which now includes all the prior art). So, the inventors who discovered that sildenafil could be orally administered to treat ED because it was a PDE V inhibitor, thought their real discovery was that PDE V inhibitors can be orally administered to treat ED, but in fact at least one was already known (epimedium aka ‘horny goat weed’ aka ‘yin yang huo’). Consequently, all they really discovered was that sildenafil can be orally administered to treat ED. If the inventors do not know what they have discovered, they might not identify the true inventive concept in the disclosure.

In my view, elaborated at more length in my post on Shire, the focus should be on the solution to the objective problem facing the inventors. In this case, the objective problem, identified in Pentney J’s discussion of the facts, was “to develop a formulation that was similarly effective [to the old Lumigan] in lowering IOP but did not cause the unwanted side effects” [17]. The question should be whether the claimed invention was an obvious solution to that problem. This didn’t make any difference on the facts, because in this case the inventors were essentially correct about what they had invented, and so the inventive concept derived from the specification was more or less the same as the objective inventive concept — similar enough to make no difference in the obviousness analysis [181]. But that will not always be true.

Pentney J also dismissed four insufficiency attacks in brief reasons. It was clear that the specification disclosed sufficient information to allow the skilled person to make and use the invention [390] and all four attacks were premised on the incorrect idea that sufficiency somehow requires more than that.

First, Juno argued that the patent failed to disclose information which would allow the skilled person to conclude it was safe for human use [386]. Pentney J pointed out the claim doesn’t say anything about the safety profile, so failure to disclose any such information is irrelevant [391]. Further, “It is well-established in the jurisprudence that the standard required to obtain a patent cannot be equated to that needed to obtain regulatory approval” [391] quoting Novo Nordisk v Cobalt 2010 FC 746 [352].

Second, Juno argued that the patent “fails to fully describe the advantage of the 0.01% dose formulations to enable the POSITA to understand its benefits over old LUMIGAN” [386]. Pentney J dismissed this by saying “I find that the specification in the 691 Patent, and in particular the data shown in the examples, is sufficient to establish that the formulation in Claim 16 would deliver comparable or better IOP reduction as compared to old LUMIGAN, and that is all that is required” [393]. I’m sure that even that much is required. The claim is simply to a formulation. The claim says nothing about its benefits, namely increased efficacy with reduced side effects. I don’t see why the specification has to establish comparable efficacy; it’s enough that the specification discloses how to make and use the invention so as to achieve whatever benefits it might have.

Third, Juno argued that the disclosure was insufficient because it did not disclose that the 200 ppm BAK served as a penetration enhancer [395]. Pentney J dismissed this on the basis that “a patent does not have to explain how the invention works as long as it explains how to work the invention” [395]. That is a straightforward application of Consolboard [1981] 1 SCR 504, which Pentney J cited.

Finally, Juno argued that “since Allergan has asserted that data based on rabbit studies cannot be directly extrapolated to the effects on humans, the Patent’s disclosures regarding permeability are not sufficient” [386]. This strikes me as another version of the third argument, arguing that the patent must disclose why the invention works. Pentney J dismissed it for similar reasons.