Promise of the Patent, Obviousness, and Overbreadth

Alcon Canada Inc v Cobalt Pharmaceuticals Co / olopatadine (NOC) 2014 FC 149 Gleason J
             2,447,924 / olopatadine / PATANOL

In Alcon v Cobalt / olopatadine (NOC), Gleason J held the 924 patent invalid for failure to satisfy the promised utility. It would be tempting to read this case as illustrating how the application of the promise doctrine has, or has not, changed since Plavix 2013 FCA 186 (blogged here). But I don’t really see this as a promise case at all. To my mind, on the facts as found by Gleason J, the claim is clearly invalid, and the more interesting question is on what ground it is invalid. I suggest that the better, or at least alternative, grounds for invalidity are that the claims at issue were obvious and overbroad.

The technical contribution underpinning the patent was the discovery that a common excipient, PVP, improves the physical stability of higher concentration olopatadine solutions [51]. On the facts, Gleason J found this insight not to be obvious [120]. As construed, the claims in issue, 2 and 7, were to olopatadine solution of a specified concentration, around the 0.2% found in PATANOL,

and PVP having an average molecular weight of 5000 [5K] to 1600K and in an amount sufficient to improve the physical stability of the solution” [47].

(The inventive concept also encompassed the fact that five other excipients, which were excluded from the claims, do not enhance stability, but this aspect of the patent was not ultimately significant to the finding of invalidity.)

The difficulty for Alcon is that while there was evidence 58K grade PVP would indeed stabilize 0.2% olopatadine solution [191], the evidence was not sufficient to either demonstrate or soundly predict that 1300K grade PVP would stabilize 0.2% olopatadine solution [197]. Consequently, Gleason J held that “the promise of utility” was not established across the entire range specified in Claim 2 [209].

The phrase “in an amount sufficient to improve the physical stability of the solution” did not help, because there was no evidence that any amount of PVP of that grade would improve physical stability. Claim 7 fell for essentially the same reason [215]. While Gleason J noted that the factual basis for a sound prediction had to be disclosed in the patent itself [125], this issue was not determinative, as there was no evidence extrinsic to the patent to show utility.

In some sense, the claimed invention had utility, in that the specified solution could be used to treat allergic and inflammatory eye reactions, even using 1300K PVP. So it seems natural to say that it lacks utility for the promised purpose, namely to stabilize the olopatadine solution. But Gleason J expressly held that the promise of the patent and its inventive concept are “one and the same” [60], [63]. Therefore we might equally say that the claimed invention lacks utility for the purpose of supporting the inventive step.

This is exactly the European position. This case is essentially the same as the famous decision of the EPO Appeal Board in T 0939/92 AgrEvo, which concerned a product claim for a class of chemical compounds useful as herbicides. While the patent sufficiently described how to make the claimed compounds, Board there was nothing inventive in doing so, and consequently, if there was any inventive step it must lie in the discovery that the compounds had herbicidal properties. Under the problem-and-solution approach to obviousness used by the EPO, the technical problem facing the inventor was to find new herbicidal compounds, and the question was whether it was obvious that the claimed compounds were the solution to this problem. But, the Board reasoned, the claimed compounds could only be an inventive solution to the problem if they were actually a solution to the problem: [2.6]. It is not obvious, in a literal sense, to say that a potion of cat hair and chocolate is a cure for cancer, but that claim is fanciful, rather than inventive, unless it is true (my example, not theirs). Consequently, the Board held that it had to be “credible” (“plausible” is now the usual term: see T 1329/04) that the compounds were herbicides, and moreover, this had to be true of “substantially all” the compounds falling within the claim [2.5.4, 2.6] Because this could not be established, the claims were held to be invalid as lacking the inventive step required by EPC Art 56.

The same line of reasoning is accepted in UK law, as was explained colorfully by Jacob LJ in Actavis v Novartis [2010] EWCA Civ 82, discussing the “5¼ inch plate paradox.”

[36] This runs like this. Suppose the patent claim is for a plate of diameter 5¼ inches. And suppose no-one can find a plate of that particular diameter in the prior art. Then (a) it is novel and (b) it is non-obvious for there is no particular reason to choose that diameter. The conclusion, that the plate is patentable, is so absurd that it cannot be so.

[37] What then is the answer to the paradox? It is this: the 5¼ inch limitation is purely arbitrary and non-technical. It solves no problem and advances the art not at all. It is not inventive. And although "inventive step" is defined as being one which is not obvious, one must always remember the purpose of that definition - to define what is inventive. That which is not inventive by any criteria is not made so by the definition. Trivial limitations, such as specifying the plate diameter, or painting a known machine blue for no technical reason are treated as obvious because they are not inventive.

So it is quite clear that in Europe, the 924 patent would have been found invalid for lack of inventive step (at least on the facts as found by Gleason J). But Gleason J held that the claimed invention was not obvious; she held that the inventive concept was that “PVP at sufficient concentrations improves the physical stability of higher concentration (0.2% to 0.6%) olopatadine solutions” [82], and that this concept was not obvious. The difference is that, implicitly at least, Gleason J did not require the claimed invention to be inventive across the full breadth of the claim. The inventive concept which she identified is that of the patent as a whole, not the inventive concept of the particular claims in issue. The inventive concept of Claim 2 in particular must be that PVP in the range of 5K to 1600K stabilizes olopatadine solutions. Suppose the claim in issue had been to 1300K PVP specifically. The European analysis would say that there is nothing inventive to a claim to 1300K PVP solution, because it does not in fact stabilize olopatadine solution. It is merely an “arbitrary” selection: T 0939/92 2.5.3. Consequently, a claim to the range including 1300K PVP is invalid on the principle that the claim must be valid across its breadth. It is clear in Canadian law that it is the inventive concept “of the claim in question,” which is at issue: Sanofi 2008 SCC 61, [67(2)], so the same analysis would imply that Claim 2 was invalid for obviousness in Canadian law as well.

The problem may also be characterized as one of overbreadth. While one way in which a claim may be overbroad is if the claim omits essential elements of the invention, the general principle is simply that “the monopoly claimed must not go beyond the consideration” Mullard Radio (1936) 53 RPC 323, 348 (HL). Straightforwardly, in this case a claim to 1300K PVP olopatadine solutions goes beyond the consideration, which is the discovery that 58K PVP olopatadine solutions has improved physical stability over what was previously known.

Gleason J did hold that the claim was overbroad, but she held that “overbreadth is simply another way of articulating the utility argument.” I am inclined to think that it is the other way around; the recent Canadian focus on utility, and the promise of the patent in particular, has led the parties and the courts characterize everything as an aspect of the promise doctrine. In the classic promise cases, the invention as claimed had sufficient utility to support a patent, and was otherwise valid, but some statement in the disclosure promised even greater utility, and that promise of greater utility was not met. In this case, the problem lies in the claim itself, which is where Gleason J found the promise at issue [55]. The real problem here is not that the specification promised more than the invention as claimed delivered; it is that the invention claimed more than the specification disclosed.

In UK law, the same problem might also be addressed as a matter of insufficiency, as discussed in Regeneron [2013] EWCA Civ 93 [100-01]:

It must therefore be possible to make a reasonable prediction the invention will work with substantially everything falling within the scope of the claim or, put another way, the assertion that the invention will work across the scope of the claim must be plausible or credible. . . .

On the other hand, if it is not possible to make such a prediction or if it is shown the prediction is wrong and the invention does not work with substantially all the products or methods falling within the scope of the claim then the scope of the monopoly will exceed the technical contribution the patentee has made to the art and the claim will be insufficient. It may also be invalid for obviousness, there being no invention in simply providing a class of products or methods which have no technically useful properties or purpose.

In context, the reference to insufficiency meant so-called “Biogen insufficiency.” I am not particularly a fan of Biogen insufficiency (and note that the EWCA says that the basis for invalidity may be either Biogen insufficiency or obviousness), as it is really another way of framing the overbreadth objection. Under the UK Patents Act 1949 the statutory basis for the overbreadth objection was the requirement in s 32(1)(i) that the be “fairly based” on the disclosure. The fair basis provision of the UK Act had been repealed by the time Biogen was decided, and As the High Court of Australia perceptively noted in Lockwood (No 1) [2004] HCA 58 [67], the House of Lords' reasoning in Biogen boils down “to the following Voltairean aphorism: ‘Since the fair basis doctrine no longer exists, it is necessary to invent it.’”

My inclination is to think that the obviousness and overbreadth approaches are most appropriate; the invention as claimed was not inventive, but arbitrary, and consequently the claim went beyond the consideration. For reasons just given, I do not see it as truly a promise case (whether or not the promise doctrine is sound), nor as a case of insufficiency, at least in the classical sense. But my point here is not to resolve this interesting conceptual debate, but simply to point out that while the promise doctrine is undoubtedly very powerful, not every defect in a patent is necessarily attributable to a false promise.

Cinar, Monsanto and Convoyed Goods

Cinar Corp v Robinson 2013 SCC 73, var’g 2011 QCCA 1361 var’g 2009 QCCS 3793

I have finally gotten around to reading through Cinar v Robinson, 2013 SCC 73, which is of interest in the patent context primarily because of its brief discussion of Monsanto v Schmeiser, 2004 SCC 34. Monsanto set out the differential profits approach to the accounting of profits remedy in patent law, and on a quick reading Cinar could be seen as suggesting that Monsanto does not state a general principled approach to an accounting, but rather a rule which applies in specific circumstances, namely when “an infringement allows the infringer to commercialize a good in a more profitable manner than he could have without the infringement” [80]. Despite this stray phrase, taken as a whole, it is clear that Cinar simply turned on the evidence, and no limitation on the differential profit approach was intended. On the contrary, Cinar reaffirms that lost profits from so-called “convoyed” goods are prima facie recoverable.

Use of Prosecution History in Claim Construction: The First Crack in the Free World Wall?

Distrimedic Inc v Dispill Inc 2013 FC 1043 de Montigny J
            2,207,045

I had somehow missed this decision last fall and was only alerted to it by the case note in the most recent IPIC bulletin. I am very glad it did not slip by me entirely, because it the case that we needed to generate a rational approach to the use of prosecution history in claim construction.

Residents of nursing homes often need a variety of different medicines to be dispensed on a weekly schedule at different times of the day. The ‘045 patent concerns a system for preparing a pill dispenser which facilitates keeping track of which pills are to be dispensed to which patient at what time of day. It consists of a plastic tray (“container-defining sheet”) with depressions to make a series of containers for holding pills to be taken four times per day (breakfast, lunch, dinner, and bedtime) over seven days [22]. After the pills are placed in the container, it is then covered by a sealing sheet on which is printed information about the prescription such as the names of the patient and the pharmacist, the date, and the medications in each container. All this was known in the prior art. It is important that the sealing sheet be properly aligned with the container, and because nursing home staff prepare many of these dispensers on a weekly basis, it is desirable to have a mechanism for aligning the container and the sealing sheet quickly and reliably. This was the problem facing the inventor [24]. The preferred solution, as described in the specification (p 3 line 13-15) was a positioning means comprising “one and preferably two upwardly projecting protuberances on the top surface of the recessed support,” and a corresponding number of “holes” in the container-sealing sheet (and see similarly p 8-9, describing the preferred embodiments).

Of course, an inventor is not confined to the most advantageous embodiment set out in the specification, and all the independent claims originally claimed a broader positioning means [206]:

Positioning means provided on at least the top surface of the container-defining sheet and on the container-sealing sheet to ensure that, in use, after the container-defining sheet is fitted onto the recessed support, the paper covering is peeled off from the bands of the container-sealing sheet and said container-sealing sheet is positioned on top of the top surface of the container-defining sheet, the bands covered with a self-adhesive material and their tearing lines be in exact superposition on top of the flanges and the dotted lines of the container-defining sheet.

These claims were originally allowed by the Patent Office, but the Notice of Allowance was withdrawn, and the claims were rejected when prior art, the Braverman Patent, was brought to the attention of the Patent Office [207].

In correspondence with the Patent Office, counsel for the applicant acknowledged that the Braverman Patent anticipated the invention as claimed, but pointed out that [208]:

BRAVERMAN does not disclose or suggest the following structural feature, which is the key feature of the present invention, namely:

d) positioning means provided on at least the top surface of the container defining sheet and on the container sealing sheet.

Such positioning means were defined in former claims 3 and 4 as being preferably protuberances and holes identified by reference numerals 5, 7 and 15 in the drawings of the present application.

Consequently, in order to overcome the objection based on the Braverman Patent, all the independent claims were amended to add a “wherein” clause specifying that the positioning means comprises

at least one upwardly projecting protuberance provided on the top surface of the recessed support, at least one hole provided into the container-defining sheet and at least one other hole provided in the container-sealing sheet, said at least one hole and one other hole being sized and positioned to correspond to and be engaged by said protuberance.

The defendant’s allegedly infringing container device had a raised outer edge of the container, and the sealing sheet was cut to fit, so that the sealing sheet was positioned by slipping it into the container. Prima facie, the defendant’s device did not use “holes” in the sealing sheet, so a key claim construction question was whether the positioning means defined by the wherein clause was an essential element of the claim.

Counsel for the defendants submitted that “it is difficult to imagine a clearer indication of the essentiality of a claim element than its addition to a claim in order to overcome an objection from the Patent Office” [208]. In my view this argument is entirely compelling. How can we allow a patentee to reclaim in litigation ground that was expressly given up in prosecution? This is the fundamental argument for the doctrine of file-wrapper estoppel, and the facts of Distrimedic powerfully illustrate the force of this argument.

Counsel for the patentee responded that prosecution history simply cannot be used in claim construction, relying on this paragraph from Free World 2000 SCC 66:

[66] In my view, those references to the inventor's intention refer to an objective manifestation of that intent in the patent claims, as interpreted by the person skilled in the art, and do not contemplate extrinsic evidence such as statements or admissions made in the course of patent prosecution. To allow such extrinsic evidence for the purpose of defining the monopoly would undermine the public notice function of the claims, and increase uncertainty as well as fuelling the already overheated engines of patent litigation. The current emphasis on purposive construction, which keeps the focus on the language of the claims, seems also to be inconsistent with opening the pandora's box of file wrapper estoppel. If significant representations are made to the Patent Office touching the scope of the claims, the Patent Office should insist where necessary on an amendment to the claims to reflect the representation.

de Montigny J responded by distinguishing Free World:

[210] While statements or admissions made in the course of patent prosecution shall not be used for the purpose of interpreting a claim, this is not what the Court is called upon to do in the case at bar. A change in the wording of a claim as a result of an objection from the Patent Office is an objective fact from which an inference may be drawn, and is not the same as representations made to the Patent Office.

There is a valid distinction between a statement or admission and a change in the wording of the claim, but, in my view, it is not enough to distinguish Free World, as statements and admissions were given only as examples of extrinsic evidence, and not an exhaustive definition. The general thrust of Free World is that extrinsic evidence and prosecution history generally are to be excluded.

However, the SCC’s holding in Free World is unpersuasive and unprincipled. The specific arguments made in [66] are unpersuasive:

[66] To allow such extrinsic evidence for the purpose of defining the monopoly would undermine the public notice function of the claims,

It would not undermine the notice function, because it operates purely as an estoppel. The worry seems to be that the public might be read the specification and claims, and believe they are safe, only to discover, after the prosecution history is consulted, that they are actually infringing. That would indeed be undesirable, because the public could not tread confidently based on the claims alone, but would have to consult the prosecution history before making any decisions. But this cannot happen. On the usual view, the prosecution history could only be used as an estoppel, preventing the patentee from reclaiming subject matter that it had given up in prosecution. The use of prosecution history can only narrow the claims, not expand them.

and increase uncertainty as well as fuelling the already overheated engines of patent litigation.

It will not increase uncertainty, because the prosecution history estoppel is not a free-standing tool. It will normally operate to clarify the meaning of claims that are uncertain on their face. In principle, the claim might be clear on its face, and then equally clear, but to the opposite effect, when the prosecution history is consulted, but this must be exceedingly rare in practice because it would mean that the examiner accepted a revision to the wording which, on its face, said the opposite of what the examiner had demanded. But in any event, even in that worst case, there would be no increase in uncertainty.

The current emphasis on purposive construction, which keeps the focus on the language of the claims, seems also to be inconsistent with opening the pandora's box of file wrapper estoppel.

It curious that Binnie J considers purposive interpretation to focus on the language of the claims, when Lord Diplock in Catnic [1982] RPC 183, 243 (HL) rejected a “purely literal” approach. The thrust of purposive construction, as compared to what went before, it to place more emphasis on the way a skilled person would read the claims. This necessarily requires going beyond the language of the claims, because a skilled person’s understanding is affected by the context of their experience in the field, their knowledge of the problems faced, and so on. Purposive construction clearly requires more extrinsic evidence, not less, than the lawyerly textual approach which it rejected. I do not see how it is inconsistent with using prosecution history. de Montigny J was quite right to say that “[a] purposive construction should obviously focus on the wording of a claim, obviously, but this is a far cry from saying that nothing else should be considered” [210]

If significant representations are made to the Patent Office touching the scope of the claims, the Patent Office should insist where necessary on an amendment to the claims to reflect the representation.

The Patent Office did insist on an amendment to reflect the representation, and the very reason we want to consult the prosecution history is to clarify the meaning of that amendment. We might always say that the Patent Office should have insisted on a clearer amendment, but the fact that textual expression is never perfect is the reason why claims construction is a crucial exercise in the first place.

The SCC’s position on the use of prosecution history estoppel is also unprincipled. As the SCC noted in Whirlpool 2000 SCC 67 [49e], a patent “is an enactment within the definition of ‘regulation’ in s. 2(1) of the Interpretation Act.” Purposive interpretation is really nothing more than the modern approach to statutory interpretation applied to the specific context of patents: see my article on “The Essential Elements Doctrine” (2011) 22 IPJ 223, 226-40. Prosecution history is the equivalent of legislative history. At one time there was an exclusionary rule against using any extrinsic aids to statutory interpretation, subject to a few narrow exceptions. That rule corresponds almost exactly to the exclusionary rule set out in Free World [66] – but that rule has long been abandoned in respect of legislation generally. (See generally Sullivan on the Construction of Statutes, 5^th ed, Ch 22.) The SCC now routinely consults legislative history in interpreting statutes. The practice is so common that citations are otiose; a quick database search will reveal multiple recent instances from the SCC alone. It is unprincipled to say that legislative history can be used to interpret legislation, and to acknowledge that patents are legislation as a matter of the Interpretation Act, and yet to refuse to use prosecution history, which is the equivalent of legislative history. Indeed, if anything, the argument in favour of considering prosecution history is even stronger than the argument for considering legislative history, as some of the traditional reasons for excluding legislative history, such as the difficulty of deciding who speaks for the legislature (see Sullivan at 595-98), have much less force in the patent context.

We should return briefly to de Montigny J’s distinction between representations and “[a] change in the wording of a claim as a result of an objection from the Patent Office [which] is an objective fact from which an inference may be drawn.” This is a sound distinction. The closest parallel in general statutory interpretation is legislative evolution, which is to say changes made in the succession of enacted texts. The use of legislative history is well established and is relatively uncontroversial, precisely because, as de Montigny J says, it is an objective fact (see Sullivan 577-78). If there is any kind of prosecution history which should be considered, this is it.

Finally, I note that the prosecution history was only one factor considered by de Montigny J in holding that the positioning means specified by the wherein clause was an essential element, and it was almost certainly not a determinative factor (see eg [211-12]). Consequently, de Montigny J’s holding regarding the use of the prosecution history could easily be distinguished by subsequent courts. It will be very interesting to see whether de Montigny J’s holding is indeed ignored, or if it is confined to changes in claim wording, or if it represents the first crack in the Free World wall, which might eventually lead to a more rational and principled approach to the use of prosecution history.

Relitigation of Same Patent Against Different Generic Survives Motion to Strike

Valeant Canada LP v Cobalt Pharma Co / diltiazem (NOC) 2013 FC 1254 Zinn J
             2,242,224 / diltiazem

In Sanofi-Aventis v Novopharm 2007 FCA 163 aff’g 2006 FC 1135 the FCA held that when an a patentee has failed in an NOC proceeding against one generic, it is an abuse of process to relitigate the same allegation of invalidity when made by second generic. In this case, Zinn J distinguished Sanofi-Aventis v Novopharm, and held that Valeant’s attempt to relitigate its ‘224 patent after having previously lost the same argument in Biovail Corp v Rhoxalpharma Inc 2005 FC 1424 (appeal dismissed for mootness 2006 FCA 92), at least survives a motion to strike.

The ‘224 patent is a formulation patent for diltiazem with a surfactant. Claim 35 and 36, which were at issue in both this litigation and Biovail, specify that the diltiazem “is released . . . with the help of a surfactant.” The question in both cases is whether, on the proper construction of these claims, the surfactant must be in the active layers, or whether it could also be in the sustained release lawyer: [10], [Biovail 28]. In Biovail Noël J held that these claims require that the surfactant be located in the active layer. In this litigation, Cobalt sought to have struck those parts of Valeant’s application which raised this same question.

Zinn J dismissed this motion, and allowed Valeant’s argument to stand. He distinguished Sanofi on the basis that “Unlike Sanofi, the present application turns on an issue of law not of fact” [22]. While this is true, this does not strike me as a particularly persuasive distinction. In Sanofi, the patentees argued that their application was not an abuse precisely because it was “a question of fact and that unlike questions of law, one court’s finding of fact is not binding on another judge considering a similar issue” [23]. Now, a prior FC decision is not strictly binding on a subsequent court at the same level, but is followed only as a matter of comity, but nonetheless, there is no evident reason why it is less abusive to relitigate a question of law than a question of fact. Certainly in Sanofi the FCA drew no such distinction: it referred generically about relitigating “the same issues” [26], and whether the allegations were “similar in all material respects” [8].

In Sanofi the FCA held that “[p]ermitting the same innovator to relitigate the same issues repeatedly poses a severe threat to the integrity of the adjudicative process,” because of the “risk of different courts reaching inconsistent results in respect of the same issues” [26], [26]. This does not strike me as entirely compelling; given that NOC proceedings are not in rem, we must accept the possibility of inconsistent results in NOC and infringement proceedings. I do not see this as a particular embarrassment, but simply a reflection of the facts that different evidence may lead to different results, and I don’t see why different factual conclusions in different NOC proceedings are any more embarrassing. With respect to questions of law, the need for legal certainty does make divergent interpretations more problematic, but I do not see the embarrassment. Questions of law may be difficult, and it may be desirable to have different interpretations tested in different proceedings before the law is finally settled. I do not think the FCA is, or should be embarrassed when it is overruled by the SCC. The law turns on the need for finality, not on the need to avoid judicial embarrassment. At the same level of court, the relevant principles are those of comity, which do not require slavish adherence to the first decision on a point of law. But whether it is persuasive or not, that was the basis for the FCA’s holding, and it applies equally in this context. Moreover, the FCA in Sanofi also pointed to the need for efficient use of judicial resources; one NOC proceeding and a subsequent infringement action is already enough litigation over a patent, without allowing multiple NOC proceedings.

Zinn J emphasized that “the blind application of the principle of consistency should not and cannot override fairness” [25]. What appears to be driving his decision is the concern that Noël J’s construction was wrong (and see also [38]), and, as just noted, comity alone is not sufficient reason to allow a wrong decision to stand. But that does not address the FCA’s point about efficiency in litigation, which implies that the patentee gets one shot at the NOC proceeding, and if it is not successful, whether rightly or wrongly, it must proceed by way of an infringement action. Zinn J rejected argument, saying “[32] Must a patentee be required to institute an infringement action or be forever foreclosed from advancing another interpretation of the claims of the patent in a future NOC proceeding? I fail to see any principled reason for adopting such a draconian position.” Whether or not there is a principled reason for such a position, it does appear to be the implication of the FCA Sanofi decision.

In any event, as Zinn J emphasized, this was only a motion to strike. Apart from the point that an argument should be clearly futile, or something close to it, in order to be struck [17-18], Zinn J held that even if this were an abuse of process, he would not exercise his discretion to strike Valeant’s argument on this issue. In part this was because on the particular facts of this case, there would be no judicial economy [38], as the ‘224 patent was one of two patents being litigated, and in Zinn J’s view the additional resources need to argue this point would be modest and could be compensated in costs. This point does go directly to the FCA’s point regarding the need for judicial economy. Zinn J’s other main point was that “I am satisfied, even considering judicial comity, that the position Valeant advances as to the interpretation of the ‘224 Patent has more than a mere possibility of success. To deny it an opportunity to present its case would be unfair” [38]. This brings us back to the points discussed above; there is a tension between justice and certainty; a strict rule that relitigation of the same issues is always an abuse comes down strongly on the side of certainty, but in Zinn J’s view, on these particular facts, the countervailing interest in justice outweighed the need for certainty and economy of judicial resources.

Construal of the Promise of the Patent, Post-Plavix, Post-Eurocopter

Pfizer Canada Inc v Mylan Pharmaceuticals ULC (NOC) 2014 FC 38 Harrington J
            2,177,576 / celecoxib / CELEBREX

The only question in this NOC proceeding was how to construe the promise of the patent. This is only the second decision since the FCA’s important Plavix, decision, 2013 FCA 186 (leave to appeal to SCC granted 30 Jan 2014) which emphasized that the promise must be explicit (blogged here), and it is the first since the FCA’s subsequent more aggressive application of the promise doctrine in Eurocopter 2013 FCA 219 (blogged here). The case is an interesting test of whether the Plavix decision marks a new attitude to the promise doctrine, as the promise language in this case was somewhat ambiguous, and the issue might have done either way. In the end, Harrington J construed the promise of the ‘576 modestly, with the result that the allegation of invalidity was not justified.

The claims of the ‘576 patent at issue claimed celecoxib itself (Claim 4), which is a COX-2 selective NSAID, and celecoxib for treating inflammation or inflammation-associated disorders (Claims 8-13) [34]. It was conceded that celecoxib was useful in the treatment of inflammation and the key question was whether the patent also promised reduced side-effects in humans [4].

The argument that the patent promised reduced side-effects turned on two main points. One is that the background to the invention, as set out in the description, is that the previously known NSAIDs, which were not COX-2 selective, caused a variety of side-effects, and so the problem faced by inventors was to reduce those side-effects [53]: “Otherwise, the patent would merely be offering another in a large class of compounds in a crowded field” [54]. The difficulty with that argument is that even if reduced side-effects is a commercial goal of the inventors, achieving the goal is not required as a matter of the law of utility: “it does not matter how crowded a field may be. If Celebrex® was new, which it was, and useful in treating inflammation, which it is, then the invention is entitled to letters patent” [66]. As the SCC put it in Consolboard [1981] 1 SCR 504, 525, “it is sufficient utility to support a patent that the invention . . . affords the public a useful choice.” Consequently, while the inventors no doubt sought a product with reduced side-effects in order develop a commercially successful product, it does not follow that they would promise to achieve that goal. While Harrington J’s reasoning on this point strikes me as entirely sound, it is in tension with Pfizer v Apotex / latanoprost (NOC) 2011 FCA 236, in which the FCA held that the patent implicitly reduced side-effects in chronic treatment of glaucoma, simply because glaucoma was a chronic disease. It seems increasingly clear that latanoprost marked a high-water mark in the aggressive interpretation of the promise.

The other main argument turned, naturally, on the wording of the patent itself. The key statements were that the discovery [28]-[32]:

(1) provides a viable target of inhibition which more effectively reduces inflammation and produces fewer and less drastic side effects

(2) Compounds of Formula I would be useful for […] the treatment of inflammation in a subject

(3) The compounds are useful as antiinflammatory agents, such as for the treatment of arthritis, with the additional benefit of having significantly less harmful side effects.

(4) Such preferred selectivity may indicate an ability to reduce the incidents of common NSAID-induced side effects.

(1) is ambiguous, as “target” may imply a goal, rather than a result [56]. (2) clearly does not imply a promise of reduced side effects. (3) was the main phrase relied on by Mylan, but this could be construed as a possible benefit, and not a promised use [57]. This is consistent even with the old leading cases on the promise doctrine: so, in Alsop’s Patent (1907), 24 RPC 733, 753 (Ch), Parker J noted that a patent would not be invalid merely because “the patentee has gone on to detail the useful purposes to which such result can be applied.” With respect to (4), Harrington J noted that “The word ‘may’ connotes a possibility; maybe yes, maybe no. While it was hoped the selectivity would reduce side effects, no such claim was made” [65]. He therefore concluded that the specification as a whole did not promise reduced side effects, and he declined to address the question of whether that putative promise was met [60].

In coming to this conclusion, Harrington J also noted that a promise should not be inferred, citing the FCA Plavix decision [68]. He also relied on the presumption set out by Zinn J in his Fournier / fenofibrate (NOC) 2012 FC 741 [126] - [127] decision (blogged here) that:

Where that promise - that claimed utility - is clearly and unequivocally expressed by the inventor in the claims of the patent, then that expression ought to be viewed as the promise of the patent. Any statement found elsewhere should be presumed to be a mere statement of advantage unless the inventor clearly and unequivocally states that it is part of the promised utility.

If, as here, the claims are certain and unambiguous in stating the promise, then the disclosure should not be examined microscopically to find additional promises that are outside the scope of the inventor’s claimed monopoly.

On its face, Zinn J’s presumption applies only when there is a utility stated in the claims. In this case the utility was expressly stated in Claims 8 - 13, but there was no utility stated in Claim 1. Harrington J cited Zinn J as saying that “a utility not expressed in the claim portion of the specification ‘[…] should be presumed to be a mere statement of advantage unless the inventor clearly and unequivocally states that it is part of the promised utility’” [70]. This seems to imply a broader presumption that any statement in the disclosure should be presumed to be a mere statement of advantage – not only when there is a utility expressed in the claims. This seems to expand on Zinn J’s statement in fenofibrate, but it is broadly consistent with the FCA’s Plavix requirement that any promise must be explicit.

Finally, Harrington J dealt with the FCA Eurocopter decision by saying that “as noted by Mr. Justice Mainville, speaking for the Court of Appeal, at para 26, the advantage ‘was principally embodied in claim 1 of the […] Patent.’ Thus, I do not see two competing schools of thought in the Court of Appeal.” [71]. If I understand correctly, Harrington J seems to be saying that Eurocopter was a case in which the promise was found in the claim itself. With respect, I do not read the claim that way. Claim 1 described the invention and embodied the key elements which produced the promised advantage (reduced ground resonance), and so in that sense it embodied the advantage, but there was no explicit reference to reduced ground resonance in the claim itself. I doubt whether Eurocopter and Plavix are really reconcilable, and given that Harrington J has followed Plavix, it is not surprising that his interpretation of Eurocopter is strained. I should say that I think that he was right to follow Plavix, because that decision had a full discussion of the correct approach to construing the promise of the patent, while the approach was only implicit in Eurocopter.

In summary, even in the pre-Plavix era, this decision might very well have turned out the same way, as the promise language at issue was relatively weak. But it might also have gone the other way, and on the whole it suggests that the Plavix decision will indeed have a moderating effect on the construal of the promise. Of course this is all subject to whatever the SCC might say in the Plavix appeal.