Obscure Prior Art Less Likely to be Combined

Google LLC v Sonos, Inc 2022 FC 1116 Zinn J

            2,545,150 / Adaptive Echo and Noise Control

This brief decision is notable for providing further clarification of how obscure prior art is to be treated in the context of an obviousness attack.

Various “virtual assistants” such as Alexa or Google Assistant, respond to voice commands that may be picked up by a range of smart appliances, such as Sonos’s well-known home audio speakers. Google’s 150 patent relates to a system of echo cancellation and noise suppression in which the order of echo cancellation and noise suppression is adaptively determined based on an amount of noise in the received signal [86]. Google brought this action alleging that Sonos’s smart speakers infringed. In defence, Sonos disputed infringement and alleged invalidity based on obviousness.

As is often the case, much turned on claim construction. A preliminary issue was whether recourse to the disclosure is always permitted or is permitted only when the claims are ambiguous. Biogen v Pharmascience 2022 FCA 143, released only a week before, settled that recourse to the disclosure is always permitted. While Zinn J would not have had the benefit of Biogen v Pharmascience, he nonetheless came to a consistent conclusion after reviewing the recent FC caselaw [42]–[44].

A key point on the facts was whether “echo cancellation” is a subset of “noise suppression” or whether the two are mutually exclusive [47]. Zinn J concluded that the two were distinct, in an analysis that turned primarily on a purposive reading of the patent as a whole, rather than on the expert evidence: see [50], noting that the two must be distinct to give meaning to Claim 7; and [51], noting that the two are treated as distinct throughout the disclosure. Zinn J’s construction of other contentious claim elements turned largely on the expert evidence.

Zinn J held that Sonos did not directly infringe because the allegedly infringing features were disabled on delivery [65], so that infringement could only be established by inducement. But he also had no difficulty in concluding that if the user infringes, that infringement will have been induced by Sonos [67]–[69], so infringement simply reduced to whether the user would infringe [70]. He answered this question in the negative, in an analysis that turned entirely on the evidence [81].

The obviousness inquiry was originally taken to be an essentially factual inquiry as to whether the invention would have been obvious to a person skilled in the relevant art. From this, it follows directly that the state of the art relevant to an obviousness attack does not comprise the entire body of publicly available information, but only that information which would actually be available to the skilled person. When the non-obviousness requirement was codified in s 28.3, it provided that obviousness is to be assessed in light of information disclosed “in such a manner that the information became available to the public in Canada or elsewhere.” On its face, this states that the entire body of public information constitutes the state of the art. The question then was whether codification was intended to change the law on this point, and if so, what the specific effect of the change was. In Hospira 2020 FCA 30, discussed here, the FCA held that “it is an error to exclude from consideration prior art that was available to the public at the relevant date simply because it would not have been located in a reasonably diligent search” but at the same time “The likelihood that a prior art reference would not have been located by a [person skilled in the art] may be relevant to consideration of step 4 of the obviousness analysis (whether differences between the state of the art and the inventive concept constitute steps which would have been obvious to the PSA) in that the uninventive PSA might not have thought to combine that prior art reference with other prior art to make the claimed invention” [86]. As I noted in my post on Hospira, this suggests that there can be no mosaicing of prior art that is not available in a reasonably diligent search. However, the FCA’s comments in Hospira were brief, as the Court remanded the issue for reconsideration.

The issue of obscure prior art arose in this case because a key element of prior art was the US patent 5,668,871. The 871 patent was very clearly obscure prior art—which is to say, prior art that would not have been found in a reasonably diligent search; it was not known to either expert prior to the litigation, nor was it found by either expert during their preparation, “despite their personal familiarity with the field and despite [Sonos’s expert] having conducted a diligent search” [92].

Google argued that the 871 patent was not eligible prior art for the purposes of the obviousness attack for that reason. Zinn J rejected this argument on the clear authority of the above-quoted paragraph 86 of Hospira [96].

The question then is how the obscure prior art may be used. Zinn J pointed out that “the difficulty of locating a document is a matter that may be considered at the final step,” relying on the same passage from Hospira [97].

[98] The question thus becomes whether, given the obscure nature of the 871 Patent, the uninventive POSITA might have thought to combine the 871 Patent with other prior art to make the claimed invention.

On the facts, Zinn J held that “given the difficulty in locating the 871 Patent, that the POSITA would not have been led directly and without difficulty to combine these references” [105], and he consequently rejected Sonos’ argument that the 150 Patent was obvious based on “any combination involving the 871 Patent” [106]. I must say that it is not clear to me exactly which references in addition to the 871 patent Zinn J was referring to as “these references” but I take it the other references were part of the prior art, but not part of the common general knowledge.

Google had relied on Kane J’s statement in Teva v Pharmascience 2020 FC 1158 [796], to the effect that “the prior art relied on by Pharmascience would not all have been found by the POSITA,” affd 2022 FCA 2 (Locke JA) [32] noting that Kane J “was apparently concerned that, given the difficulty in locating certain prior art, the PSA would not have been led directly and without difficulty to combine these references.” Zinn J considered that Locke JA was not contradicting his own prior statement in Hospira but was simply saying that “the difficulty of locating a document is a matter that may be considered at the final step” [97]. I think this is right, considering Kane J’s statement at [801] that “I agree with Teva that Pharmascience has not explained how or why the POSITA would regard these pieces of prior art together in a mosaic to lead to the invention,” and moreover “even if the POSITA was handed this mosaic, it does not lead them directly to the subject matter of the claims.” So, I don’t see any inconsistency.

Zinn J’s decision is consistent with Fothergill J’s decision in dTechs 2021 FC 190 (see here). However, it is possible that it diverges from Manson J’s decision in Biogen v Taro 2020 FC 621; in my post on Manson J’s decision I suggested that his analysis was notable for making a mosaic of obscure prior art. Manson J’s decision holding the patent at issue invalid for obviousness was just affirmed by the FCA in Biogen v Pharmascience 2022 FCA 143 [143]–[171]. However, the FCA decision did not address this issue; the main ground of appeal was that Manson J had failed to adopt the mantle of the POSITA, essentially by failing to recognize that the skilled person in the field would be highly skeptical of any new alternative treatments that were not supported by double-blind, placebo-controlled trials [144], [148]. So far as I can see, the issue of mosaicing of obscure prior art was simply not addressed at all. I think the most likely explanation is that I was simply wrong in thinking that the prior art relied on by Manson J was obscure prior art, which would explain why the issue was not raised on appeal. In that case, there is in fact no divergence between these decisions on this issue. In any event, even if Manson J did combine obscure prior art, the fact that the FCA affirmed on obviousness cannot be taken as holding that it is permissible to combine obscure prior art, given that the issue was not discussed. If the FCA had intended any comment on that point, it would surely have discussed Hospira.

Allergan v Apotex / Risedronate Dosage Form — Overview

Allergan Inc v Apotex Inc 2022 FC 260 Kane J

2,602,188 / risedronate / ACTONEL DR

This very long decision—262 pages and 946 paragraphs—turns largely on its facts. The length of the decision is a consequence of Kane J’s very thorough recitation of the arguments and evidence. The main point of interest arises from the somewhat unusual claim.

Allergan’s 188 patent relates to a dosage form of a bisphosphonate for use in treating osteoporosis. Bisphosphonates were well known for use in treating osteoporosis. Bisphosphonates were known to have poor bioavailability. One specific known problem was significantly lower absorption when taken with food—the “food effect”. The food effect was thought to be due to bisphosphonates forming complexes with ions, such as calcium and magnesium, from the food in the stomach. (Bisphosphonates were known to be absorbed in the small intestine.) The standard way of addressing the food effect in the prior art was to prescribe the bisphosphonate to be taken while fasting. This gave rise to compliance problems [273].

Allergan came up with a solution to the food effect, namely a dosage form comprising a combination of bisphosphonate with EDTA in an enteric coating. The enteric coating allows everything to pass into the small intestine, where the EDTA—a known chelating agent—binds to the cations and prevents them from binding to the bisphosphonate. This means that the dosage would be equally effective whether given with food or after fasting. Accordingly, Claim 1 of the 188 patent was to “An oral dosage form of a bisphosphonate for use with or without food or beverage intake, comprising...” All the claims incorporated the “for use with or without food” limitation. The “with or without” food structure of the claim is a bit odd, as “or” is normally used to signal alternatives, so that it might have been separated into two claims, one claiming use with food and one without. But given that the purpose of the invention was to overcome the food effect, the experts agreed, and Kane J held, that the claim should be construed to mean that the oral dosage can used “either with food or without food or beverage at the preference of the patient” [228]. That seems right to me.

But there’s a catch. Giving EDTA without food can cause irritation of the esophageal lining. Even though this was apparently an uncommon and fairly minor problem, and did not affect efficacy in treating osteoporosis, the product monograph for Allergan’s product, ACTONEL DR, consequently specified that it should be taken with food. The Apotex PM was essentially identical [905]. Consequently, Kane J held that Apotex’s product did not infringe, either directly or by inducement, because it was not for use with or without food, but only with food [921].

Kane J’s construction seems right, given the phrase “for use with or without food” in the claim. But I am puzzled as to why the patentee included those words in the first place. Kane J said this:

[225] For the ‘188 Patent, the ability to use the oral dosage form either with food or without food is set out in the claim because its purpose is to overcome the food effect (which is the problem of the extremely low absorption of the bisphosphonate when taken with food as opposed to when taken fasted). The oral dosage can be taken either with food or without food at the choice of the patient and the treatment effect will be similar (i.e., pharmaceutically effective absorption). The invention of the ‘188 Patent signalled a significant change in the administration of oral bisphosphonates, hence the emphasis on this feature.

It’s true that the advance was to allow a formulation that can be taken without food as well as with it, and it may be that is what motivated the applicant to include this phrase. But that does not mean that this feature has to be included in the claim itself. (I should note that this explanation was not central to Kane J’s reasoning, as it was not disputed that “with or without food” was an essential element.) So long as the compound itself is new, the purpose need not be specified: if an inventor comes up with a new drug that is useful for curing cancer, they are entitled to claim the compound per se, without restricting it to the use in curing cancer. In this case, the claimed composition* was new: it was essentially a selection over the closest prior art, a prior Brazilian patent application referred to as BR 601: see eg [366], [372]. So I’m at a bit of a loss to explain why it was included.** If the phrase was indeed unnecessary, that suggests the possibility that “for use without or without food” is not an essential element, but this argument was not run by Allergan [188]. So I presume there was some reason for including the phrase that is not apparent from the decision. I wonder if the phrase might have been added to placate an obstinate examiner? I note that even if it was added in response to an objection, it is not clear that this precludes a finding of non-essentiality: see Actavis [2017] UKSC 48 UKSC.

Allegan also argued that “for use with or without food” should be construed as meaning only ‘suitable for’, not intended for. Kane J rejected this argument, which I’ll discuss at more length in my next post.

In light of this construction, infringement turned on the factual issue as to whether the PM instructed use with food, or whether taking with food was optional. Allergan’s argument was to the effect that a skilled person would recognize that it would be effective if taken without food, since the side effects were minor. Nonetheless, as Kane J found [905]–[922], the PM clearly and repeated instructed that the product should be taken with food: eg ““Do not take ACTONEL DR before food or on an empty stomach. . .” That this instruction was to avoid abdominal pain rather than to enable effective absorption does not change the fact that the product was for use with food. Apotex’s PM was the same, and so Apotex did not infringe directly or by inducement.

There are a couple of other points to note. One of the expert witnesses had been ‘blinded’, but Kane J was not impressed: “I note that the jurisprudence is mixed on the treatment of blinded evidence. I favour the approach noted in Janssen Inc v Apotex Inc, 2019 FC 1355 at paras 58-59 . . . that blinded opinions are not necessarily given greater weight just because they are blinded.”

Another point is that the most relevant prior art, a prior Brazilian patent application referred to as BR 601, was obscure prior art; that is, it was available to the public, but it would not have been found by a skilled person in a reasonably diligent search [634]. This raised on the facts the holding in Hospira 2020 FCA 30 that the state of the art for the purposes of an obviousness attack includes all prior art. In my post on Hospira, I suggested that this holding leaves open the question of whether obscure prior art can be mosaiced in an obviousness attack.

Kane J noted that

[635] In accordance with Hospira at para 86, it would be an error for the Court to exclude BR 601 from the mosaic of prior art because it would not be found by the skilled person in a reasonably diligent search.

[636] As guided by Hospira, the likelihood that BR 601 would not turn up (and, based on the evidence, it would not turn up) is relevant to the obvious to try test; i.e., would the skilled person have thought to combine BR 601 with the other prior art to bridge the differences between the state of the art and the invention?

She then noted that

[637] This is a tricky distinction when applied to the facts. If BR 601 is part of the prior art, then the differences between the state of the art and the invention would need to account for whatever can be distilled from BR 601 as part of the “state of the art”. However, if the skilled person would not have found BR 601, how can the skilled person consider it in the context of combining it with other prior art at the obvious to try stage of the analysis of obviousness? This is only possible in circumstances where the unknown prior art is handed to the skilled person. To avoid this dilemma and running afoul of Hospira, I have considered BR 601 as if it were prior art, but as noted, the “teaching” of BR 601 requires careful scrutiny.

Ultimately, Kane J did not have to wrestle too directly with this “tricky distinction”, because, as I read it, BR 601 simply would not make the 188 patent obvious, no matter how it was used. BR 601 was at best a broad genus than encompassed the claimed invention, but did not point to it with any specificity, and did not give any hint of the special advantages of the 188 dosage form.

*The original version of the post said the "compound" was new: that is not correct. It is the claimed composition that was new. 

**Since the composition was new, it seems to me that it could have been claimed as such, without that phrase, in the same way a new compound can be claimed without specifying the use. 

Making a Mosaic of Obscure Prior Art Is Permissible in Obviousness Attack under Hospira

Biogen Canada Inc v Taro Pharmaceuticals Inc 2020 FC 621 Manson J

            2,562,277 / fampidrine sustained release / FAMPYRA / NOC action

As outlined in my last post, the 277 patent, owned by Acorda and licenced to Biogen, relates to the use of sustained release fampidrine for improving walking in a person with multiple sclerosis (MS). In this decision Manson J held the 277 patent, to be anticipated and obvious, ultimately because the inventors had themselves disclosed the invention in a financial filing prior to the claim date. My last post dealt with anticipation. This post deals with obviousness. This case is the first clear application of the FCA’s holding in Hospira that the state of the art for the purpose of an obviousness attack includes all prior art, not just that which would be reasonably discoverable. For convenience, I’ll refer to prior art that is not part of the cgk or reasonably discoverable, as “obscure prior art.” Importantly, Manson J appears to have been of the view that it is permissible to mosaic various pieces of obscure prior art.

The prior art included a financial filing referred to as Acorda S-1, that disclosed results from a small clinical trial undertaken by Acorda, the MS-F201 trial, as well as the protocol for a second trial that was planned (the MS-F202 trial) [36, 122]. This was held to anticipate. It was also invoked in support of obviousness, along with the Goodman References (also disclosing the same trials), and a publication reporting the pharmacokinetic properties of fampridine SR (“Hayes”). Given the very small gap between this prior art and claimed invention, the holding that the claims were obvious in light of these references followed fairly directly [181-200].

The key issue on obviousness was not bridging the gap between the prior art and the claimed invention, but whether the prior art references were properly part of the state of the art. In particular, it was not established that any of these references would have been part of the common general knowledge or discoverable in a reasonably diligent search. As Manson J noted, prior to codification of the non-obviousness requirement in s 28.3, “citable prior art for obviousness was limited to that which would have been found by the POSITA in a reasonably diligent search” [152]. However, in Hospira 2020 FCA 30 the FCA held that “it is an error to exclude prior art from the obviousness analysis simply because it would not have been located in a reasonably diligent search” [153]. Biogen objected to this as being contrary to the intent of s 28.3, but as Manson J pointed out, “The Federal Court of Appeal’s decision in Hospira is both clear and binding on this Court. Prior art will not be excluded from the obviousness analysis solely because the POSITA would not have found it after a reasonably diligent search” [154]. Consequently, Manson J considered the prior art at issue to be part of the state of the art for an obviousness attack, without consideration of whether it would have been reasonably discoverable [163-64, 166].

That much follows directly from Hospira. But there was an important point left open in Hospira, where the FCA stated (my emphasis):

[86] The likelihood that a prior art reference would not have been located by a PSA may be relevant to consideration of step 4 of the obviousness analysis (whether differences between the state of the art and the inventive concept constitute steps which would have been obvious to the PSA) in that the uninventive PSA might not have thought to combine that prior art reference with other prior art to make the claimed invention.

In my post on Hospira, I said the following:

This indicates that there can be no mosaicing of prior art that is not available in a reasonably diligent search. Thus, obviousness can be assessed over the common general knowledge alone, or over one item of broader prior art plus cgk, but not over two items of broader public art plus cgk. I believe that this is essentially the UK position.

In other words, I read Hospira as saying that while one element of obscure prior art could be combined with the traditional state of the art (cgk or material reasonably discoverable), it was not necessarily permissible to combine two elements of obscure prior art, on the view that a POSITA who did not know about a piece of obscure prior art would not think to combine it with another piece of obscure prior art which she also did not know about.

However, Manson J did not read this paragraph the same way. He stated

[193] Following the Federal Court of Appeal’s guidance in Hospira that it may be relevant to consider the likelihood that a prior art reference would not have been located by the POSITA at this stage of the obviousness analysis, I am satisfied that the POSITA would have combined the pharmacokinetic parameters taught in Hayes 2003 with the fixed doses of 10 mg bid taught in the Acorda S-1. Both references report information on the Elan fampridine SR composition, and the Acorda S-1 states that the pharmacokinetic characteristics of fampridine SR in subjects with MS had been established in earlier trials sponsored by Elan.

There was no suggestion that either Hayes 2003 of Acorda S-1 were part of the cgk or were reasonably discoverable: in light of Manson J’s holding that it was wrong to exclude prior art from the obviousness analysis for that reason, he simply never addressed the question.

Consequently, in Biogen v Taro, Manson J did mosaic two elements of obscure prior art. He was evidently of the view that the only inquiry mandated by Hospira [86] is whether the POSITA, with both elements of prior art in hand, would find it obvious to combine them.

Manson J’s interpretation of Hospira [86] is certainly reasonable, but I like to think my interpretation is reasonable as well, so it’s not clear, at least to me, what the FCA intended. Presumably the point will be clarified by the FCA soon enough. If Manson J’s reading is correct, then the state of the art available for an obviousness attack in Canadian law now goes beyond what is available in either the US or the UK (at least as I understand the law in those jurisdictions.)

The State of the Art Includes All Prior Art

Hospira Healthcare Corporation v. Kennedy Trust for Rheumatology Research 2020 FCA 30 Locke JA: Rivoalen, Nadon JJA var’g 2018 FC 259 Phelan J
            2,261,630 / infliximab / INFLECTRA

The third step of the Windsurfing/ Pozzoli approach to obviousness, as approved by the Supreme Court in Sanofi 2008 SCC 61 [67], requires the court to “[i]dentify what, if any, differences exist between the matter cited as forming part of the ‘state of the art’ and the inventive concept of the claim or the claim as construed.” But what is the “state of the art”? The FCA in this case has clarified a point which has been obscure since codification of the non-obviousness requirement a quarter of a century ago.

The requirement that a patentable invention must involve an inventive step first emerged in the English case law in the late nineteenth century. Originally conceived of as an aspect of novelty, it soon because recognized as a distinct requirement, often referred to as requiring proper “subject-matter.” The current requirement that the invention be non-obvious to a person of ordinary skill in the art originated as a test for inventiveness on the particular facts of the case, propounded by Sir Sanford Cripps as counsel in Sharp & Dohme v Boots Pure Drug Co (1928), 45 RPC 153 (CA) at 162-63 and adopted by Lord Hanworth MR at 173. So conceived, the obviousness inquiry was taken to be an essentially factual inquiry as to whether the invention would have been obvious to a person skilled in the relevant art. From this it follows directly that the state of the art does not comprise the entire body of publicly available information, but is restricted to that information which would actually be available to the skilled person. So, in General Tire [1972] RPC 457, 497-98, the English Court of Appeal considered the state of the art to include only that material that would be available to a skilled person after a reasonably diligent search.

This was well settled law prior to codification. A question then arose on codification because s 28.3 of the Act provides that obviousness is to be assessed in light of information disclosed “in such a manner that the information became available to the public in Canada or elsewhere.” On its face this states that the entire body of public information constitutes the state of the art. The operative wording is exactly the same as that in the novelty provision (though the novelty provision, in a separate paragraph, also includes prior patent applications, including unpublished applications, in order to implement the first-to-file priority regime). Thus, on the face of the Act, the state of the art for purposes of obviousness corresponds exactly to the prior art that is relevant to anticipation, excluding prior unpublished patent applications.

The issue arose in this case because Phelan J had excluded two items of prior art from consideration at step 3 of the Windsurfing / Pozzoli test on the basis that they would not have been located in a reasonably diligent search [FC 214]. The FCA held that Phelan J had erred on this point.

Locke JA reviewed the Canadian case law, and noted that while a number of cases, including some affirmed by the FCA, had indeed excluded some elements of prior art, that point had never been specifically affirmed or even commented on by the FCA, and the point therefore could not be considered to be settled. My sense is that notwithstanding the text of the act, the Canadian courts have generally operated under the assumption that the codification was not intended to substantively change the law, and it is because of that underlying assumption that courts had not previously been pressed for a decision. Locke JA also noted that some decisions, including Ciba v SNF 2017 FCA 225 (see here) had suggested that the state of the art was broader in scope. However, none of these cases had definitively resolved the issue.

Locke JA also noted at [85] that “commentaries by authors well-known in the field of patent law have expressed doubt as to whether it is appropriate to limit the scope of relevant prior art to the results of a diligent search since the wording of section 28.3 of the Patent Act is not so limited.” I must say that I am disappointed that my own short article on this topic, “What is the State of the Art for the Purpose of an Obviousness Attack?” (2012) 27 CIPR Review 385-94, was not cited. In that article I pointed out that exactly that question was before the House of Lords in Technograph [1972] RPC 346 and English Court of Appeal in General Tire and Windsurfing [1985] RPC 59 in interpreting the parallel English legislation, which raised the same issue. Lord Reid and Lord Diplock took opposing views in Technograph, and the Court of Appeal in General Tire preferred Lord Reid’s view, but sided with Lord Diplock in Windsurfing. The point was expressly obiter in all these cases, and the point was never definitively resolved in English law prior to the adoption of the EPC with the Patents Act, 1977. The division of opinion in these decisions illustrates that the text itself is not determinative and it is unfortunate that the FCA did not devote more time to discussing these cases, which presumably were not cited to it. (While I did not come to a firm conclusion in my article, I had argued that the context of the enactment, in particular the history of the relevant English decisions on which Canadian law was based, suggested that the codification was not intended to alter the state of the art, but was only intended to establish the relevant dates for assessing obviousness, in particular to implement the grace period.)

Locke JA then addressed the substantive issue. I’ll set out the sentences of the key paragraph one at a time:

[86] In light of section 28.3 of the Patent Act and the applicable jurisprudence and commentaries, I conclude that it is an error to exclude from consideration prior art that was available to the public at the relevant date simply because it would not have been located in a reasonably diligent search.

This is clear: all prior art must be consided under step 3 of the obviousness analysis.

[86] The likelihood that a prior art reference would not have been located by a PSA may be relevant to consideration of step 4 of the obviousness analysis (whether differences between the state of the art and the inventive concept constitute steps which would have been obvious to the PSA) in that the uninventive PSA might not have thought to combine that prior art reference with other prior art to make the claimed invention.

This indicates that there can be no mosaicing of prior art that is not available in a reasonably diligent search. Thus, obviousness can be assessed over the common general knowledge alone, or over one item of broader prior art plus cgk, but not over two items of broader public art plus cgk. I believe that this is essentially the UK position.

Locke JA concluded with a policy point:

[86] However, excluding prior art simply because it is difficult to find is problematic because it would result in the possibility of a valid patent on an invention that had, but for some non-inventive tweak, already been disclosed to the public. In my view, that is not what Canada’s patent regime is intended to permit.

With respect, I do not find this persuasive. It is well-established that the mere fact that an item of prior art is “available to the public” does not mean that it is ever likely to be of any actual benefit to the public. “It is settled law that there is no need to prove that anybody actually saw the disclosure provided the relevant disclosure was in public. Thus an anticipating description in a book will invalidate a patent if the book is on a shelf of a library open to the public, whether or not anybody read the book and whether or not it was situated in a dark and dusty corner of the library.” Lux Traffic Controls [1993] RPC 107 at 133, quoted with approval in Baker Petrolite 2002 FCA 158 [42.6] (emphasis in Baker Petrolite). If there is prior art that would have made the invention obvious to a skilled person, but that would not be located in a reasonably diligent search, then it would only be located as part of an extended or arduous search. An invention may be found to be non-obvious because it would require experimentation that is “not routine, but rather was prolonged and arduous” Sanofi [89]. If that is true for experimentation, it is not clear why it should not be true for searching prior art. The traditional view is that the obviousness inquiry asks whether a person skilled in the relevant art would arrive directly at the invention when presented with the relevant problem. If the answer is no, I don’t see why it should matter whether that is the answer because discovering the invention would require prolonged and ardous technical experiments, or a prolonged and arduous prior art search plus routine experimentation.

On a related point, as noted above, the traditional inquiry was essentially factual, albeit focused on an objective skilled person, rather than any particular individual. The “state of the art” as being what would be discovered in a reasonably diligent search reflects what a skilled person would actually do when addressing the problem. If the state of the art includes all prior art, then we have moved away from this factual inquiry, into an odd hybrid, asking whether a skilled person would find the invention obvious in light of information which would not actually have been be available to them.

In any event, regardless of the policy issue, Locke JA’s conclusion is certainly supported by the text of the Act, and the point now seems to be settled.

What is the State of the Art for the Purpose of an Obviousness Attack?

Aux Sable Liquid Products LP v JL Energy Transportation Inc 2019 FC 581 Southcott J
            2,205,670

The invention at issue in Aux Sable v JL Energy is relatively straightforward, but the decision raises several interesting legal issues. One of the key points is a pure question of law: what qualifies as part of the state of the art for the purpose of an obviousness attack? As it happens, I wrote an article on this topic a few years ago, "What is the State of the Art for the Purpose of an Obviousness Attack?” (2012) 27 CIPR 385, though it was not cited in the decision.

The main attack on Claims 1-8 was based on obviousness. The third step of the Windsurfing / Pozzoli test, as adopted in Sanofi 2008 SCC 61, [67] asks what differences, if any, “exist between the matter cited as forming part of the ‘state of the art’ and the inventive concept of the claim or the claim as construed.” The question then, is what constitutes the “state of the art”? Traditionally there was a distinction between the prior art which could be raised in an attack based on anticipation, and the state of the art which might form the basis of an obviousness attack. The prior art comprises all subject matter which was available to the public at the relevant date, while the state of the art was a subset of the prior art, comprising only that which was generally known to a person skilled in the art, or which would be discovered by a skilled person in a reasonably diligent search prior to addressing the problem at hand (the “reasonably diligent search test” [144]): and see eg 2006 FC 1234 [57] aff’d 2007 FCA 217 [37]; 2009 FCA 222 [73]. However, the amendment which came into effect in 1996, codifying the obviousness requirement in s 28.3, describes the subject matter which can be raised in an obviousness attack in exactly the same terms as for anticipation in s 28.2, namely information disclosed “in such a manner that the information became available to the public in Canada or elsewhere.” The question is whether this codification was intended to change the law, so that the state of the art for the purposes of an obviousness attack is now co-extensive with the prior art available for anticipation. Aux Sable, the accused infringer, and the plaintiff in this impeachment action, argued that state of the art is the same as the prior art [136], while the patentee, JL Energy, argued that the reasonably diligent search test applied in determining the state of the art for purposes of obviousness [137].

Aux Sable made a straightforward argument. The plain meaning of the text of s 28.3 – “in such a manner that the information became available to the public in Canada or elsewhere” – captures the prior art, without qualification by the reasonably diligent search test [147]. Further, it was common ground that s 28.2 does not require that a prior art reference be locatable by a reasonable diligent search in order to be citable for an anticipation allegation, and since the language is the same in both sections, this implies the test must be the same [147], [150]. In reply JL Energy, relied on jurisprudence post-dating the amendments, which it argued continued to endorse or apply the reasonably diligent search test. Southcott J carefully reviewed the case law [152]-[174], and concluded, in my view correctly, that it provided little assistance either way, generally because the point was not fully argued, or the decision was ambiguous (see the synopsis below). Southcott J therefore concluded that:

[176] I find Aux Sable’s statutory interpretation arguments compelling, and I disagree with JL Energy’s response that this issue has been settled in its favour by existing jurisprudence.

He therefore held that “a prior art reference which was disclosed to the public, prior to the applicable date prescribed by s 28.3, forms part of the prior art for purposes of an obviousness analysis, regardless of whether the reference would have been locatable through a reasonably diligent search” [176].

This analysis is sound so far as it goes. The text of the Act undoubtedly supports Southcott J’s conclusion, and while the text is not determinative of the correct interpretation, it is dominant. Consequently, a strong argument based on legislative history, policy or established case law would be required to overcome the plain meaning of the text: Canada Trustco Mortgage Co 2005 SCC 54 [10]; Biolyse Pharma 2003 FCA 180 [13].

I would suggest, however, that there is a strong argument, or at least one worth considering, based on the legislative history and policy, which supports the view that the codification was not intended to change the law. I made an argument to that effect in my article “What is the State of the Art for the Purpose of an Obviousness Attack?” (2012) 27 CIPR 385. This article was not referred to by Southcott J, presumably because it was not cited to him by JL Energy. Perhaps it was not the kind of reference that would be found in a reasonably diligent search?

I won’t go through the entire article, but I will address one point. Southcott J suggested, at [174], that:

a patent monopoly should not result from an obvious advancement upon prior art, regardless of whether the prior art would have been located in a diligent search. . . . JL Energy has not offered any policy argument as to why the locatability of a piece of prior art should matter for purposes of obviousness if it does not matter for anticipation.

I suggest that there is a relevant difference between obviousness and anticipation. In Graham v John Deere (1966) 383 US 1 at 11, the US Supreme Court explained the justification for the non-obviousness requirement as follows:

[T]he underlying policy of the patent system [is] that "the things which are worth to the public the embarrassment of an exclusive patent" ... must outweigh the restrictive effect of the limited patent monopoly. The inherent problem was to develop some means of weeding out those inventions which would not be disclosed or devised but for the inducement of a patent.

That is, an invention should be considered obvious if it would have been developed even without the lure of a patent, and conversely, it should be considered non-obvious if the patent reward was required to bring it forth.

The obviousness implements this “but for” approach on the view that the lure of a patent would not be required if the invention could easily be made by a person of ordinary skill in the art when faced with the problem at hand. It has therefore traditionally been understood as an essentially factual inquiry as to whether the invention would have been obvious to a person skilled in the relevant art. As an essentially factual inquiry, the state of the art for the purposes of obviousness must be that which an actual skilled person would have had at hand. The factual nature of the inquiry is reflected in the famous Cripps question, formulated by Sir Stafford Cripps in argument in Sharp & Dohme v Boots Pure Drug Co (1928), 45 RPC 153 (CA), at 162-63 and adopted by Lord Hanworth MR at 173, which is the basis of the test set out in the Act:

"The real question is: was it for all practical purposes obvious to any skilled chemist in the state of chemical knowledge existing at the date of the patent which consists of the chemical literature available (a selection of which appears in the Particulars of Objections) and his general chemical knowledge, that he could manufacture valuable therapeutic agents by [the process specified in the patent]

Anticipation aims at the same purpose, to weed out patents for inventions which did require the lure of the patent monopoly, but the question is in some ways simpler. Instead of asking whether a skilled person would have invented it without the patent lure, we simply ask whether it has already been made public; if so, the lure of a patent is evidently not required to deliver the knowledge to the world.

My article also discusses various points of legislative history, suggesting that these also point in the same direction. Whether such considerations are sufficient to displace the prima facie interpretation based on the text of the Act is a different issue. I must admit that I am disappointed that Southcott J did not have these arguments before him.

After all that, Southcott J considered all the references advanced by Aux Sable, including those which would not have been found in a reasonably diligent search, and he nonetheless held the claims at issue to be valid. Thus the holding that the state of the art includes the entire prior art did not ultimately affect the result.

Here is a synopsis of Southcott J’s review of the cases. Starting with the FCA cases, JL Energy sought to rely on a passage from Mishan 2015 FCA 163 [18]-[23], but Southcott J noted that the FCA at [21] had expressly left that question open, and it could not be taken as having settled the issue [155]. Southcott J then considered Ciba 2017 FCA 225. I had interpreted the majority (Pelletier and Rennie JJA) as holding that all prior art that is available in a novelty attack is also part of the state of the art for the purposes of an obviousness attack [51]-[63], though Woods JA dissented, on the view that “[t]he issue is better addressed in an appeal where it is relevant to the outcome and in which the Court has the benefit of full submissions from counsel, which was not the case here” [156]. I noted in my post on Ciba that I agreed entirely with Woods JA, and I am pleased to see that her dissent was given significant weight by Southcott J in holding that Ciba had not settled the point either, though he also discussed significant ambiguity in the interpretation of the majority decision [157-60].

Southcott J then reviewed a number of post-1996 cases which referenced or applied the reasonably diligent search test, but he agreed with Aux Sable’s position that the issue had not been squarely raised, and so were not strong authority for the proposition that the amendment had changed the law [161]-[164]. This strikes me as sound. He then noted that while Novartis 2015 FC 770 supported JL Energy’s position, it had considered only case law, and not the statutory interpretation arguments [166]. Southcott J found more comfort in Locke J’s decision in Pollard 2016 FC 883, which I described as being “clearly sympathetic to the argument that the state of the art should comprise all public prior art,” but Southcott J recognized that Locke J had canvassed the arguments [170], and while it supported Aux Sable’s position, it was more a matter of raising some persuasive considerations, than a determinative decision [174].

The State of the Art on State of the Art?

Ciba Specialty Chemicals Water Treatments Ltd v SNF Inc 2017 FCA 225 Pelletier JA: Rennie JA, concurring reasons by Woods JA aff’g 2015 FC 997 Phelan J
            2,515,581

Under the “old” Patent Act the body of prior art that may be set up against the patent in an obviousness attack did not include all publicly available information that could be set up in a novelty attack, but only that which was generally known to a person skilled in the art or which would be discovered in a reasonably diligent search. As Hughes J noted in Merck v Pharmascience 2010 FC 510, [37], there is a “quaere” as to whether codification of the obviousness requirement in s 28.3 of the new Act, which requires a person skilled in the art to have regard to information “available to the public” as of the claim date, has changed the law in this respect, so that all publicly available prior art may be used in an obviousness attack, regardless of whether it would have been discovered by a reasonably diligent search. In my article, What is the State of the Art for the Purpose of an Obviousness Attack? (2012) 27 CIPR 385, I reviewed the debate. While the text of 28.3 does imply that all publicly available prior art may be set up against the patent in an obviousness attack under the new Act, I concluded, based on contextual and purposive considerations, that on balance the better view is that s. 28.3 was not intended to change the law.

In Ciba Specialty Chemicals Water Treatments Ltd v SNF Pelletier JA, with Rennie JA agreeing, held that all prior art that is available in a novelty attack is also part of the state of the art for the purposes of an obviousness attack [51]-[63]. This implies that the codification did indeed change the law, though Pelletier JA did not say so expressly. Woods JA concurred in the result, but declined to express an opinion on this point [99], saying “The issue is better addressed in an appeal where it is relevant to the outcome and in which the Court has the benefit of full submissions from counsel, which was not the case here” [100]. I must say that I agree entirely with Woods JA. While the position taken by Pelletier JA is a reasonable one which is supported by the text of s 28.3, some important interpretive arguments were not addressed, presumably because the Court did not have the benefit of full submissions on this point.

Pelletier JA relied primarily on the text of 28.3 [51], [57], which does indeed clearly support his position. The remainder of the points made by Pelletier JA are, in my view, more questionable.

Pelletier JA stated that “In Anglo-Canadian jurisprudence, the comparison at step 3 is to the prior art” [51]. Pelletier JA cited no cases on point, and it may be that Pelletier JA did not intend this as a description of what the courts had previously done, but rather a statement of his conclusion. To the extent that he intended it as describing the actual practice of the courts, I am not sure it is right. As discussed in my article, it was not true in English law prior to the Act of 1977 and it was not true in Canadian law prior to 1993. Since then, the question has remained unsettled in Canadian law. In 2006 FC 1234, [109] Hughes J expressed the view, albeit in passing, that codification was not intended to change the law. In Novartis v Teva / deferasirox 2015 FC 770 blogged here, O’Reilly J held that indeed the law had not changed, while in Pollard v BABN 2016 FC 883 [192-95], Locke J suggested the opposite. The question was noted, but not addressed, in 2015 FCA 163 [21]: see here for a discussion. None of these cases were discussed by Pelletier JA. And indeed, as Pelletier JA acknowledged [59], he himself had expressed the opposite view in BMS v Teva 2017 FCA 76, [48]. This is not to say that BMS v Teva settled the question, as it was not directly at issue, and also referred to “prior art” in other paragraph, as Pelletier JA pointed out. But it does indicate that the point was open.

Pelletier JA also relied on Windsurfing [1985] RPC 59 (CA) for the proposition that the state of the art encompasses all prior art. Windsurfing does indeed take that position, but, as discussed in my article, it was in obiter and in conflict with the well known decision in General Tire [1972] RPC 457, 497 (CA).

He also relied at [56] on the SCC’s use of the phrase “prior art” in Sanofi / Plavix 2008 SCC 61, [70]-[71]. However, the issue there was not the precise definition of the prior art, but rather the significance of the actual course of conduct of the inventor.

Finally, Pelletier JA relied on the parallel between s 28.3 and the legislation under the UK Patent Act, 1977 [58]. In my article, at 389, I argued that the better parallel is the UK Act of 1949, both because the structure of the 1977 Act is different, and because it was widely recognized as effecting a change in the law, whereas 28.3 was thought to be to be merely a codification.

None of this is to say that Pelletier JA's conclusion was wrong, but only that it is not clearly right. (Note that the above is not an exhaustive review of the relevant considerations which are discussed in my article, but only touches on those raised by Pelletier JA.)

Despite the clear holding by Pelletier JA, I am not sure where the law stands on this point. The authority of the holding is undermined by the fact that it was a decision of only two members of the panel, the point was not fully argued, it did not review the relevant Canadian cases, and it was strictly obiter. In Lilly v Apotex / cefaclor, 2009 FC 991 [349]-[370], Gauthier J declined, quite rightly in my view, to follow the SCC holding as to “standard of review” of a granted patent set out in Wellcome / AZT 2002 SCC 77 [44], in part because the issue was not argued before the SCC [356, fn126] and the point was obiter [357]. (With that said, the situations are not entirely parallel because the SCC in Wellcome / AZT was clearly wrong, for the reasons given by Gauthier J, while the position taken by Pelletier JA is entirely reasonable, even though the counter-arguments were not fully considered.) Apart from any question of the authority of this decision, it does seem to me undesirable for a court to make a holding in obiter on a point of law which has not been argued, because, as Wellcome / AZT illustrates, such holdings may not be as well considered as when the issue is fully argued.

TRUVADA Patent Invalidated by Conference Call With Investors Outside of Grace Period

Gilead Sciences, Inc v Apotex, Inc 2016 FC 856 Brown J
            2,512,475 / TRUVADA

In this chapter of the continuing TRUVADA saga, Brown J held Gilead’s 475 patent invalid as being anticipated, or in the alternative, obvious, on the basis of a disclosure made by Gilead executives in the course of a conference call with investors. The filing date was 13 January, 2004, and the conference call, which was public [67], took place more than one year before, on 2 December, 2002, and so outside the one year grace period for disclosure by the inventor set out in s 28.2(1)(a) and s 28.3(a). The argument for anticipation or obviousness based on the conference call was very strong, and the main tactic of Gilead’s counsel was to try to keep the conference call transcript out of evidence by instructing the Gilead witnesses not to produce the conference call transcript, in violation of Rule 94(1), and without seeking the relief of Rule 94(2) [60], [61]. Brown J held that in the circumstances hearsay evidence of the conference call transcript introduced by Apotex was admissible as being both necessary and reliable [60]. The holding on anticipation [97] and obviousness [120-21] followed directly. Apotex also argued lack of utility, but Brown J construed the promise of the patent modestly [139], and found that the promised utility was soundly predicted [143].

As a minor point, Brown J found that the conference call, which was public [67], would have been known to a person skilled in the art [103], but I don’t see a specific holding by Brown J that it was part of the state of the art, in the sense of being common general knowledge or prior art which would be discovered in a reasonably diligent search directed to the problem at hand. As noted here, there is an issue as to whether the art that can be used in an obviousness attack includes only the state of the art, so defined, or all prior art. This would potentially have been an important point had the decision rested primarily on obviousness, but given that anticipation was the primary basis for Brown J’s holding of invalidity, nothing turned on it.

Comments on the State of the Art for an Obviousness Attack

Pollard Banknote Limited v BABN Technologies Corp 2016 FC 883 Locke J
            2,752,551

“What is the State of the Art for the Purpose of an Obviousness Attack?” That’s the title of an article I wrote in 2012 (27 CIPR 385). The puzzle is this. Traditionally, under the old Act, the body of prior art which could be set up against the patent in an obviousness attack (the “state of the art”), included only the common general knowledge plus prior art which would be discovered in a reasonably diligent search directed to the problem at hand. But s 28.3, which codified the obviousness in the new Act, provides that the relevant information is that “disclosed before the claim date … in such a manner that the information became available to the public”. That includes all public prior art, not just prior art which would have been found in a reasonably diligent search. There is an open question as to whether s 28.3, which was generally thought to have simply codified the existing non-obviousness requirement, had extended the state of the art. In my article I concluded that “contextual and purposive considerations imply that s. 28.3 was not intended to change the law, and the state of the art remains limited to information that is reasonably discoverable,” but “the argument based on the text of the provision is sufficiently strong that case law directly addressing the question will be needed before the point can be considered settled.”

While a number of decisions under the new Act simply assumed that the relevant state of the art was the same as under the old Act, the first decision to address this question directly, albeit briefly, was Novartis v Teva / deferasirox 2015 FC 770 blogged here, in which O’Reilly J held that indeed the law had not changed. (The question was noted, but not addressed, in 2015 FCA 163 [21].) Now, in Pollard v BABN, Locke J has suggested that he may prefer the opposite conclusion [192-95]. The question was relevant in respect of the Camarato application, which was crucial to the holding that Claim 1 was obvious. As I read his decision, Locke J never directly held that the state of the art extends to all publicly available prior art. Locke J noted the problem and made some comments on it, but his only direct holding was “Even if I was of the view that this reference’s findability in a diligent search was a relevant consideration, I would nevertheless conclude that it is citable for obviousness,” because it was indeed have been found in a diligent search, and so would be part of the state of the art under either test [196]. The phrase “even if I was” is ambiguous. It might suggest that he was of the view and findability is not the test, and his holding that the Camarato application was findable was in the alternative; but it might also mean that it didn’t matter whether findability was the test, given that the reference was in fact findable.

Regardless of the precise holding, Locke J was clearly sympathetic to the argument that the state of the art should comprise all public prior art:

[194] [I]n order to obtain a valid patent, it is not enough for a skilled person simply to make an obvious change to what is known in the art. This principle should apply to any information that was available to the public, even if it would not have been located in a diligent search. For example, should a skilled person be able to obtain a valid patent by simply searching a dusty corner of a public library for a document that describes a forgotten invention and making an obvious change to it?

The problem with the final rhetorical question is that even under the traditional test a person cannot obtain a patent by “simply” searching a dusty corner and making an obvious change, because if the search was simple, by definition it would have been found in a reasonably diligent search. The proper question is whether a skilled person should be able to obtain a valid patent by making an obvious modification to a document that was found after a long and arduous search. The answer to that question is not so easy. The EWCA in Windsurfing [1985] RPC 59, 71 made a very similar argument, which I discuss at more length in my article at 391-93.

State of the Art for an Obviousness Attack Includes Only Art Discoverable in a Reasonably Diligent Search

Novartis Pharmaceuticals Canada Inc v Teva Canada Ltd 2015 FC 770 O'Reilly J
            2,255,951 / deferasirox / EXJADE

O’Reilly J’s decision in Novartis v Teva / deferasirox does not break wholly new legal ground, but it does develop several interesting points regarding the state of the art in obviousness, Viagra-style insufficiency, and the promise of the patent. I will deal with these issues in separate posts.

The ‘951 patent relates to iron chelators, which are used to treat disorders caused by excess iron in the body [7]. Iron chelators were known in the prior art, but the most effective could not be administered orally, and others were toxic or insufficiently effective [9]. The problem facing the inventors was therefore to find an effective iron chelator that could be orally administered. Two classes of compounds are disclosed in the patent: Formula I compounds were previously known, and the relevant claims were to their use for treatment of diseases involving excess iron in humans, while Formula II compounds were novel, and they were claimed both as compounds and for the use in treatment of iron excess diseases [14]. Deferasirox, the active ingredient in EXJADE, is a Formula II compound and so is covered by both use and compound claims.

The obviousness attack turned primarily on the facts, but a pure question of law was raised as to what constitutes the prior art for the purposes of an obviousness attack. O’Reilly J held that there was considerable doubt as to whether some of the sources formed part of the relevant prior art for the purposes of an obviousness attack because they would not have been located by a skilled person conducting a reasonably diligent search. But section 28.3 of the Act provides that an invention must not be obvious having regard to information that “became available to the public” in Canada or elsewhere before the relevant date. This is exactly the same language as used in the novelty provision, s 28.2. Citing Barrigar’s, Canadian Patent Act Annotated, at PA-341, Teva argued that this means that the old requirement that only reasonably discoverable information can be used in an obviousness attack is no longer the law, and the prior art for obviousness and novelty purposes should be the same [53]. 

Teva has a good argument based on the text of the Act, but I have argued that “contextual and purposive considerations imply that s. 28.3 was not intended to change the law, and the state of the art remains limited to information that is reasonably discoverable”: What is the State of the Art for the Purpose of an Obviousness Attack?, (2012) 27 CIPR 385 at 394. I went on to note that “the argument based on the text of the provision is sufficiently strong that case law directly addressing the question will be needed before the point can be considered settled.” In this case, O’Reilly J stated that “there is case law applying the usual ‘reasonably diligent search’ criterion even after the enactment of s 28.3. I see no reason to take a different approach here” [53]. While O’Reilly J is right to say that there was prior case law using the traditional criterion, the question of whether s 28.3 had modified the law was not raised in those decisions. O’Reilly J’s considered holding is therefore a significant development on this point of law. Strictly, however, his holding was obiter, since he held that the invention was not obvious even if Teva’s interpretation was correct [54]. With that said, I believe that O’Reilly J's decision is the first to directly address this point, and it certainly reinforces the view that properly interpreted, s 28.3 did not change the law.

The State of the Art for the Purposes of Obviousness

This is a plug for my most recent article, “What is the State of the Art for the Purpose of an Obviousness Attack?” which has just appeared in 27(2) CIPR 385. Traditionally the body of prior art which may be set up against a patent in an obviousness attack includes only that prior art which is common general knowledge or reasonably discoverable by a person skilled in the art. However, the codification of the obviousness requirement in s 28.3 refers to information disclosed “in such a manner that the information became available to the public in Canada or elsewhere,” which is almost exactly the same wording as used in respect of anticipation, in s 28.2. On its face this implies that the entire body of public information is available in respect of an obviousness attack, as it is for an argument of anticipation: see AD Morrow, “Patent Update” (2003), 20 CIPR 237 at 248 suggesting that in consequence of the codification, the prior law has been “swept away, and everything available to the public is now to be considered.” Less dramatically, as Hughes J noted in Merck & Co Inc v Pharmascience Inc 2010 FC 510 [37], there is a “quaere” as to whether the codification of the non-obviousness requirement in section 28.3 of the new Act has changed the law, so that all publically available prior art may be used in an obviousness attack, regardless of whether it would have been discovered by a reasonably diligent search. My article reviews the debate, including the English case law, which addressed exactly the same question, inconclusively, under the Patents Act, 1949. In considering the purpose of the provision, I argue that the English Court of Appeal in Windsurfing v Tabur Marine [1985] RPC 59 – the source of the Windsurfing / Pozzoli test approach to obviousness – was wrong in principle when it held that the prior public, but isolated, use of a sailboard by Peter Chilvers, a 12 year old boy, was part of the state of the art for purposes of an obviousness attack. I conclude that the best view is that the codification has not changed the law, but the textual argument is strong enough that a judicial decision will be required to settle the question definitively.