Eli Lilly Canada Inc v Nu-Pharm Inc / olanzapine 2011 FC 255 Snider J
When a generic pharmaceutical company obtains an NOC for a drug, it seems natural to assume that it intend to begin selling that drug, and patentee pharma companies are often anxious to begin infringement proceedings as soon as possible. The courts have been consistently unwilling to allow an action to proceed solely on the basis that the generic has obtained an NOC: AstraZeneca Canada Inc. v Novopharm Ltd. / rosuvastatin, 2009 FC 1209 Hughes J. affm’d 2010 FCA112 Noël JA: Pelletier, Dawson JJA, is the leading case, both for Hughes J’s review of the case law, and because his decision striking the Statement of Claim was affirmed by the FCA. The basic problem for the patentee is that the Federal Court has been unwilling to infer that the generic will enter the market prior to the expiry of the relevant patent, simply from the fact that it has obtained an NOC (see 2009 FC 1209 [14]). A quia timet claim for future infringement will fail for this reason [ibid 23]. A bare assertion that the defendant has manufactured the compound etc. will be disregarded as too speculative to support a claim of current infringement if made without any evidentiary foundation [ibid 18].
Snider J's decision in the olanzapine case confirms and extends this line of cases. In the rosuvastatin case, the generic had not obtained an NOC at all, which added an layer of speculation, particularly as the patentee was seeking an order of prohibition at the same time that it sought to launch an infringement action. In the olanzapine case, Nu-Pharm had obtained an NOC, days after the patent had been declared invalid in infringement proceedings (Eli Lilly Canada Inc. v. Novopharm Ltd. / olanzapine 2009 FC 1018) and before the trial decision invalidating the patent was reversed and remanded (2010 FCA 197). However, Snider J nonetheless held that the Statement of Claim should be struck, as there was no allegation of activity outside of the regulatory safe-harbor of section 55.2(1). Snider J made this holding on the assumption that the preparation of the ANDS and the obtaining of the NOC are in themselves acts of infringement [25], a point which is not yet settled.
Beyond these details, the case affirmed the general point that a patentee cannot launch an infringement action merely because the generic has obtained an NOC: “the pleadings disclose nothing beyond an assertion that Nu-Pharm is positioning itself, through an unnamed third party, to enter the market for olanzapine and, that by doing so, Nu-Pharm will infringe the patent. . . .[T]he Statement of Claim is, at least in part, very much a quia timet proceeding to which the findings of Justice Hughes and the Court of Appeal in AstraZeneca FCA are applicable” [31].
Direct evidence of commercial stockpiling would presumably suffice. But what about the patentee’s basic point that one can reasonably infer an intent to infringe from the fact that the generic has obtained an NOC? (While the question of whether obtaining an NOC is itself an act of infringement is in principle open, I cannot see how getting permission to make or sell is the same as making or selling.) It would be interesting to have statistical evidence on this point. If it turned out that a generic that obtained an NOC only launched during the term of the relevant patent 30% the time, then the view that obtaining an NOC cannot support a quia timet action would seem sound. But if that figure is 95%, the common sense inference would be stronger. Of course, it is not clear whether a court would find statistical evidence relevant to the intent of a particular defendant, and in any event I am not aware of any such study. I’d certainly be curious to find out what those stats are.
Tuesday, March 29, 2011
Monday, March 28, 2011
Disgorgement of Profits under Section 8 in the Ontario Courts
Apotex Inc. v. Abbott Laboratories, Ltd. / lansoprazole 2010 ONSC 6909, 89 CPR(4th) 141, Whitaker J
Section 8 of the PM(NOC) Regulations provides that if a patentee is unsuccessful in its application for an order of prohibition (as to what is “success” see here), the generic is entitled to compensation for having been kept out of market in consequence of the statutory stay. The generic would obviously prefer that compensation to be measured by the patentee’s profits rather than by its own loss, as the patentee’s profits with market exclusivity will be greater than the profits the generic would have made. However, the FCA has held that an accounting of profits is not available under section: Merck Frosst Canada Ltd v Apotex Inc / alendronate (NOC) 2009 FCA 187 [89] affm’g 2008 FC 1185. This was under an older version of the provision, which provided for “relief by way of damages or profits” (s 8(4)). The same conclusion presumably follows under the current provision, which omits the words “or profits.”
In Apotex v Syntex / naproxen 2005 FC 480 [22], von Finckenstein J said that the NOC Regulations are a “complete code,” which suggests there is no alternative route by which a generic can get disgorgement of the patentee’s profits. However, on appeal the FCA, while repeating that section 8 is “a complete code for the recovery of damages by a second person against a first person,” 2005 FCA 424 [9], at the same time suggested that “while strictly speaking it might be said there is no such tort as negligent breach of a statute, there nevertheless can be a claim in negligence against the Crown and proof of a statutory breach that causes damages may be evidence of such negligence” [10]. This seems to imply that while section 8 is a complete code in respect of losses flowing from the statutory stay as such, if there is some other basis for liability, section 8 may not govern. This leaves open the door to the possibility that a claim for unjust enrichment could allow the generic to claim disgorgement of profits rather than damages. However, it seems to be established that such a claim cannot be made in the Federal Court, as a claim that is not based on the Regulations would not be within the jurisdiction of the court: 2009 FC 693 [8].
In June 2009 Apotex obtained an NOC for apo-lansoprazole, and it commenced an action in the Federal Court for compensation under section 8. Just at that time, the FCA released its the alendronate decision, holding that an accounting is not available under section 8. Apotex discontinued its Federal Court action [28] and decided to try its luck in the Ontario courts, which do of course have jurisdiction over a claim in unjust enrichment. This strategy has cleared the first hurdle. In Apotex Inc. v Abbott Laboratories, Ltd. 2010 ONSC 6909 Abbott moved to strike Apotex’s claim for disgorgement of profits based on unjust enrichment. Whitaker J dismissed the motion, holding that it is not “plain and obvious” that such a claim could not succeed. This seems to me to be the correct conclusion given the stringent test on a motion to strike, but it of course remains to be seen how this claim will fare on the merits.
I suggest that the 2006 amendment of section 8(4) indicates a legislative intent to confine the generic to damages generally – if the legislature intended the generic to be able to recover the patentee’s profits, surely it would have been simplest to allow this to be done under the Regulations themselves, in the Federal Court. However, if the Ontario courts conclude that a successful generic should be entitled to a disgorgement of profits, as a matter of fairness, this argument is unlikely to be conclusive.
The question of whether the successful generic should be entitled to a disgorgement is an interesting one. A prohibition order consequent on PM(NOC) proceeding is analogous to an interlocutory injunction, and section 8 is analogous to the undertaking required of the party who obtains such an injunction: 2008 FC 1185 [54]. Such undertakings do not generally require a disgorgement of profits. I am not aware of any Canadian cases addressing the question of whether it would be appropriate to require an accounting on an undertaking the context of an interlocutory injunction, but the point has been raised in a recent UK patent case. In Wake Forest University Health Sciences v. Smith & Nephew [2009] EWHC 45 (Pat), Lewison J declined to require an undertaking of that form, saying “it is by no means clear to me that it would be just to transfer a profit made by the claimant to the enjoined defendant. If the claimant has made a profit which it would not have made but for the injunction, there may be other people to whom it would be more just that those profits should be returned, either other potential competitors with the defendant or customers who, as things turn out, may have been overcharged” [19]. It will be interesting to see whether the Ontario courts find this reasoning persuasive.
A last general point. While there is an analogy between the NOC proceedings and an interlocutory injunction, one basic distinction is that the PM(NOC) proceedings are entirely separate from any subsequent infringement action. This question of whether section 8 can give rise to a separate cause of action illustrates once again the difficulties arising from a separation of the statutory stay from the underlying cause of action. For illustrations of other problems arising from this same cause, see here and here.
PS - This blog normally comments on recent cases; while this case was released last fall, it has only come to my attention recently, which is why I am commenting on it now.
Section 8 of the PM(NOC) Regulations provides that if a patentee is unsuccessful in its application for an order of prohibition (as to what is “success” see here), the generic is entitled to compensation for having been kept out of market in consequence of the statutory stay. The generic would obviously prefer that compensation to be measured by the patentee’s profits rather than by its own loss, as the patentee’s profits with market exclusivity will be greater than the profits the generic would have made. However, the FCA has held that an accounting of profits is not available under section: Merck Frosst Canada Ltd v Apotex Inc / alendronate (NOC) 2009 FCA 187 [89] affm’g 2008 FC 1185. This was under an older version of the provision, which provided for “relief by way of damages or profits” (s 8(4)). The same conclusion presumably follows under the current provision, which omits the words “or profits.”
In Apotex v Syntex / naproxen 2005 FC 480 [22], von Finckenstein J said that the NOC Regulations are a “complete code,” which suggests there is no alternative route by which a generic can get disgorgement of the patentee’s profits. However, on appeal the FCA, while repeating that section 8 is “a complete code for the recovery of damages by a second person against a first person,” 2005 FCA 424 [9], at the same time suggested that “while strictly speaking it might be said there is no such tort as negligent breach of a statute, there nevertheless can be a claim in negligence against the Crown and proof of a statutory breach that causes damages may be evidence of such negligence” [10]. This seems to imply that while section 8 is a complete code in respect of losses flowing from the statutory stay as such, if there is some other basis for liability, section 8 may not govern. This leaves open the door to the possibility that a claim for unjust enrichment could allow the generic to claim disgorgement of profits rather than damages. However, it seems to be established that such a claim cannot be made in the Federal Court, as a claim that is not based on the Regulations would not be within the jurisdiction of the court: 2009 FC 693 [8].
In June 2009 Apotex obtained an NOC for apo-lansoprazole, and it commenced an action in the Federal Court for compensation under section 8. Just at that time, the FCA released its the alendronate decision, holding that an accounting is not available under section 8. Apotex discontinued its Federal Court action [28] and decided to try its luck in the Ontario courts, which do of course have jurisdiction over a claim in unjust enrichment. This strategy has cleared the first hurdle. In Apotex Inc. v Abbott Laboratories, Ltd. 2010 ONSC 6909 Abbott moved to strike Apotex’s claim for disgorgement of profits based on unjust enrichment. Whitaker J dismissed the motion, holding that it is not “plain and obvious” that such a claim could not succeed. This seems to me to be the correct conclusion given the stringent test on a motion to strike, but it of course remains to be seen how this claim will fare on the merits.
I suggest that the 2006 amendment of section 8(4) indicates a legislative intent to confine the generic to damages generally – if the legislature intended the generic to be able to recover the patentee’s profits, surely it would have been simplest to allow this to be done under the Regulations themselves, in the Federal Court. However, if the Ontario courts conclude that a successful generic should be entitled to a disgorgement of profits, as a matter of fairness, this argument is unlikely to be conclusive.
The question of whether the successful generic should be entitled to a disgorgement is an interesting one. A prohibition order consequent on PM(NOC) proceeding is analogous to an interlocutory injunction, and section 8 is analogous to the undertaking required of the party who obtains such an injunction: 2008 FC 1185 [54]. Such undertakings do not generally require a disgorgement of profits. I am not aware of any Canadian cases addressing the question of whether it would be appropriate to require an accounting on an undertaking the context of an interlocutory injunction, but the point has been raised in a recent UK patent case. In Wake Forest University Health Sciences v. Smith & Nephew [2009] EWHC 45 (Pat), Lewison J declined to require an undertaking of that form, saying “it is by no means clear to me that it would be just to transfer a profit made by the claimant to the enjoined defendant. If the claimant has made a profit which it would not have made but for the injunction, there may be other people to whom it would be more just that those profits should be returned, either other potential competitors with the defendant or customers who, as things turn out, may have been overcharged” [19]. It will be interesting to see whether the Ontario courts find this reasoning persuasive.
A last general point. While there is an analogy between the NOC proceedings and an interlocutory injunction, one basic distinction is that the PM(NOC) proceedings are entirely separate from any subsequent infringement action. This question of whether section 8 can give rise to a separate cause of action illustrates once again the difficulties arising from a separation of the statutory stay from the underlying cause of action. For illustrations of other problems arising from this same cause, see here and here.
PS - This blog normally comments on recent cases; while this case was released last fall, it has only come to my attention recently, which is why I am commenting on it now.
Friday, March 25, 2011
Latanoprost decision affirmed
Pfizer Canada Inc. v. Pharmascience Inc. / latanoprost (NOC), 2011 FCA 102 Noël JA: Trudel, Mainville JJA affm’g 2009 FC 1294 Heneghan J
The opinion of the FCA affirming Heneghan J’s decision granting an order of prohibition respecting Latanoprost ophthalmic solution is of course of interest to the parties, but there is little in it of general legal interest. This is not surprising, given the cautious style of Heneghan J’s decision. She set out the law by quoting leading cases, reviewed the evidence descriptively, and then set out her conclusions briefly in light of that evidence. While counsel for the appellant did their best to find an appealable point of law, in the end the appeal turned primarily on findings of fact. As there was enough evidence to support her conclusions, the appeal was dismissed.
The opinion of the FCA affirming Heneghan J’s decision granting an order of prohibition respecting Latanoprost ophthalmic solution is of course of interest to the parties, but there is little in it of general legal interest. This is not surprising, given the cautious style of Heneghan J’s decision. She set out the law by quoting leading cases, reviewed the evidence descriptively, and then set out her conclusions briefly in light of that evidence. While counsel for the appellant did their best to find an appealable point of law, in the end the appeal turned primarily on findings of fact. As there was enough evidence to support her conclusions, the appeal was dismissed.
Wednesday, March 23, 2011
Interest is very . . . interesting
Eli Lilly Co. v. Apotex Inc. 2009 FC 991, 80 C.P.R. (4th) 1, aff'd 2010 FCA 240.
Since I don't have any current cases to blog about, this post reflects on an aspect of one of last year's more interesting decisions. Interest was traditionally not awarded on damages at law, as it was thought to be punitive. A fortiori, compound interest was not permitted. This has been statutorily remedied in each province to the extent of permitting interest to be awarded, both pre- and post-judgment. The Federal Courts Act s 36-37 provides that if a cause of action arises solely in one province (i.e. infringement occurs entirely and only in one province), the laws of that province govern the award of interest, and otherwise the provisions of the Federal Courts Act are applicable. Section 36(2) gives the Court the discretion to award pre-judgment interest “at any rate that the Federal Court of Appeal or the Federal Court considers reasonable in the circumstances.” The Court of Appeal in Wellcome / AZT [2001] 1 FC 495 at [123] has said that “the exercise of discretion in awarding interest "must be related to the task of putting the plaintiff in the same position, so far as money is concerned, as he would have been if he had not suffered the loss.” However, the Federal Court not uncommonly specifies pre-judgment interest at the annual average Bank of Canada bank rate – the rate of interest the Bank of Canada charges on one-day loans to major financial institutions – without consideration of whether this rate would in fact provide full compensation: see Merck & Co., Inc. v. Apotex Inc. 2006 FC 524 [240] affm’d 2006 FCA 323; Laboratoires Servier v. Apotex Inc. 2008 FC 825, [513] affm’d 2009 FCA 222. If the successful plaintiff is not able to borrow at the bank rate, then such an award would arguably not put it in the position it would have been in had it not suffered the loss in the first place.
Another difficulty with the statutory scheme is that many of the Acts, including both the Federal Courts Act, and the Ontario Courts of Justice Act, which is substantially similar, prohibit compound interest pre-judgment. The Federal Courts Act, s 36(4)(b) provides that "[i]nterest shall not be awarded under subsection (1), (b) on interest accruing under this section." (The Onario Act, ss 128(4)(b) is in the same words.) It is clear that an award of simple interest will not provide full compensation to a successful party, particularly when the time between the infringement and judgment is long, or when prevailing interest rates are high.
Since I don't have any current cases to blog about, this post reflects on an aspect of one of last year's more interesting decisions. Interest was traditionally not awarded on damages at law, as it was thought to be punitive. A fortiori, compound interest was not permitted. This has been statutorily remedied in each province to the extent of permitting interest to be awarded, both pre- and post-judgment. The Federal Courts Act s 36-37 provides that if a cause of action arises solely in one province (i.e. infringement occurs entirely and only in one province), the laws of that province govern the award of interest, and otherwise the provisions of the Federal Courts Act are applicable. Section 36(2) gives the Court the discretion to award pre-judgment interest “at any rate that the Federal Court of Appeal or the Federal Court considers reasonable in the circumstances.” The Court of Appeal in Wellcome / AZT [2001] 1 FC 495 at [123] has said that “the exercise of discretion in awarding interest "must be related to the task of putting the plaintiff in the same position, so far as money is concerned, as he would have been if he had not suffered the loss.” However, the Federal Court not uncommonly specifies pre-judgment interest at the annual average Bank of Canada bank rate – the rate of interest the Bank of Canada charges on one-day loans to major financial institutions – without consideration of whether this rate would in fact provide full compensation: see Merck & Co., Inc. v. Apotex Inc. 2006 FC 524 [240] affm’d 2006 FCA 323; Laboratoires Servier v. Apotex Inc. 2008 FC 825, [513] affm’d 2009 FCA 222. If the successful plaintiff is not able to borrow at the bank rate, then such an award would arguably not put it in the position it would have been in had it not suffered the loss in the first place.
Another difficulty with the statutory scheme is that many of the Acts, including both the Federal Courts Act, and the Ontario Courts of Justice Act, which is substantially similar, prohibit compound interest pre-judgment. The Federal Courts Act, s 36(4)(b) provides that "[i]nterest shall not be awarded under subsection (1), (b) on interest accruing under this section." (The Onario Act, ss 128(4)(b) is in the same words.) It is clear that an award of simple interest will not provide full compensation to a successful party, particularly when the time between the infringement and judgment is long, or when prevailing interest rates are high.
Monday, March 21, 2011
Slavery and Gene Patents
Harvard College v Canada (Commissioner of Patents), 2002 SCC 76, [2002] 4 SCR 45
Patentable subject matter is a hot topic these days, from Bilski in the USSC, to the reference regarding Programs for Computers G 3/08, in the EPO, to the decision of Phelan J in Amazon.com 2010 FC 1011, which is now on its way to the FCA. Here is a shameless plug: I have three papers on this topic. In the first, I argue that this area of the law has suffered from the conflation of two distinct doctrines: the rule against abstract claims, which prohibits abstract claims in any field of endeavour; and field-specific exclusions, which prohibit patents in a particular field, regardless of whether the claim is abstract or applied. From this distinction I develop a general framework for approaching the issue, and I then apply that framework to argue that Phelan was right for the right reasons. The second paper looks at the history and principles behind the rule against abstract claims. The third is a criticial analysis of the US jurisprudence, in which I argue that the USSC has pretty much gotten it all wrong, and while Canadian law generally is in much better shape, Schlumberger, which followed the USSC, was wrongly decided.
Back to the topic of this post. In Harvard Mouse the majority in the Supreme Court held higher life forms to be unpatentable subject matter. The majority’s view was that a variety of concerns unique to higher life forms meant that patents are not “the appropriate vehicle” for protecting this type of invention [120]. The majority raised a variety of objections, but for the most part these were not particularly powerful, in the sense that even if one accepts their validity, they did not seem to present enough of a problem to warrant a wholesale denial of patents in a productive new field of technology. The majority’s argument was that all these objections cumulatively indicated patenting of higher life forms just doesn’t fit well with the Patent Act. Perhaps this was the majority’s real concern, but my sense is that one argument in particular tilted the balance. This is the “[t]he potential for commodification of human life” [176], which the majority termed “the most significant issue addressed by the [Canadian Biotechnology Advisory Committee]” [175].
Patentable subject matter is a hot topic these days, from Bilski in the USSC, to the reference regarding Programs for Computers G 3/08, in the EPO, to the decision of Phelan J in Amazon.com 2010 FC 1011, which is now on its way to the FCA. Here is a shameless plug: I have three papers on this topic. In the first, I argue that this area of the law has suffered from the conflation of two distinct doctrines: the rule against abstract claims, which prohibits abstract claims in any field of endeavour; and field-specific exclusions, which prohibit patents in a particular field, regardless of whether the claim is abstract or applied. From this distinction I develop a general framework for approaching the issue, and I then apply that framework to argue that Phelan was right for the right reasons. The second paper looks at the history and principles behind the rule against abstract claims. The third is a criticial analysis of the US jurisprudence, in which I argue that the USSC has pretty much gotten it all wrong, and while Canadian law generally is in much better shape, Schlumberger, which followed the USSC, was wrongly decided.
Back to the topic of this post. In Harvard Mouse the majority in the Supreme Court held higher life forms to be unpatentable subject matter. The majority’s view was that a variety of concerns unique to higher life forms meant that patents are not “the appropriate vehicle” for protecting this type of invention [120]. The majority raised a variety of objections, but for the most part these were not particularly powerful, in the sense that even if one accepts their validity, they did not seem to present enough of a problem to warrant a wholesale denial of patents in a productive new field of technology. The majority’s argument was that all these objections cumulatively indicated patenting of higher life forms just doesn’t fit well with the Patent Act. Perhaps this was the majority’s real concern, but my sense is that one argument in particular tilted the balance. This is the “[t]he potential for commodification of human life” [176], which the majority termed “the most significant issue addressed by the [Canadian Biotechnology Advisory Committee]” [175].
Thursday, March 17, 2011
The Faint Hope of the Essential Elements Doctrine
Bauer Hockey Corp. v. Easton Sports Canada Inc., 2011 FCA 83 Noël JA: Létourneau, Trudel JJA aff’g 2010 FC 361, 83 CPR(4th) 315 Gauthier J
The decision of the FCA in Bauer v Easton illustrates the deficiencies of the essential elements doctrine in claims construction.1 In principle the doctrine can sometimes rescue a patentee from a poorly drafted patent, but in this case it did not, and in general it is extremely difficult to predict how it will be applied. I suggest that it needs to abandoned entirely, or reinvigorated with a clarified version of the Improver questions, as discussed in my article, “The Essential Elements Doctrine in Patent Infringement" 22 IPJ 223, (draft version here.) As it stands, the doctrine adds complexity and reduces certainty, without offering more than a faint hope to a patentee with imperfect claims.
The doctrine is important when a claim includes an element that is not necessary to the patentability of the disclosed invention. If the element in question is “essential” a competitor can copy the inventive concept without infringing; conversely, if the problematic element is not “essential,” a defendant who copies all of the remaining essential elements will infringe the claim, even though its embodiment does not include the element in question. The dilemma in such cases is that if the court finds the element to be non-essential, it is in effect re-writing the claim by ignoring that element in determining infringement, but if it finds the element to be essential, it is allowing the competitor to freely take an invention that was invented and disclosed by the patentee. There is no easy way out of this dilemma. The Improver questions purported to offer a test, but that approach has not been applied in Canadian cases, despite its endorsement by the Supreme Court in Free World 2000 SCC 66 [55]-[56], and in the UK they have been repudiated by Lord Hoffmann himself as a “mangle” in Kirin-Amgen [2004] UKHL 46 [52].
The decision of the FCA in Bauer v Easton illustrates the deficiencies of the essential elements doctrine in claims construction.1 In principle the doctrine can sometimes rescue a patentee from a poorly drafted patent, but in this case it did not, and in general it is extremely difficult to predict how it will be applied. I suggest that it needs to abandoned entirely, or reinvigorated with a clarified version of the Improver questions, as discussed in my article, “The Essential Elements Doctrine in Patent Infringement" 22 IPJ 223, (draft version here.) As it stands, the doctrine adds complexity and reduces certainty, without offering more than a faint hope to a patentee with imperfect claims.
The doctrine is important when a claim includes an element that is not necessary to the patentability of the disclosed invention. If the element in question is “essential” a competitor can copy the inventive concept without infringing; conversely, if the problematic element is not “essential,” a defendant who copies all of the remaining essential elements will infringe the claim, even though its embodiment does not include the element in question. The dilemma in such cases is that if the court finds the element to be non-essential, it is in effect re-writing the claim by ignoring that element in determining infringement, but if it finds the element to be essential, it is allowing the competitor to freely take an invention that was invented and disclosed by the patentee. There is no easy way out of this dilemma. The Improver questions purported to offer a test, but that approach has not been applied in Canadian cases, despite its endorsement by the Supreme Court in Free World 2000 SCC 66 [55]-[56], and in the UK they have been repudiated by Lord Hoffmann himself as a “mangle” in Kirin-Amgen [2004] UKHL 46 [52].
Monday, March 14, 2011
What is the Promise of a Patent?
GlaxoSmithKline Inc. v Pharmascience Inc. / rosiglitazone (NOC) 2011 FC 239, Hughes J
1,328,452 / rosiglitazone / AVANDIA
In Canadian law it is standard to say that “[w]here the specification does not promise a specific result, no particular level of utility is required; a ‘mere scintilla’ of utility will suffice. However, where the specification sets out an explicit "promise", utility will be measured against that promise” Eli Lilly Canada / olanzapine 2010 FCA 197 [76]. The notion that the utility is measured by the promise of the patent is now peculiar to Canadian law. It came to us from English law, but it is no longer a consideration in that country: Pharmacia Corp. v Merck & Co. Inc. [2001] EWCA Civ 1610 [54]. While it was raised most commonly in the utility context, it was based on the provision of the earlier Act that a patent might be revoked if “the patent was obtained on a false suggestion or representation” (see Raleigh Cycle v Miller [1948] All ER 308 (HL), 324). Both that provision as a ground of revocation, and the requirement of utility as such, were abolished by the Patents Act, 1977. The doctrine seems never to have been part of US law.
The result of the strict application of the doctrine may be that an inventor with a perfectly good invention will find its patent invalid because the invention’s virtues were exaggerated in the disclosure. This result is rare, as the courts are generally unwilling to read the promise so broadly as to invalidate a patent for an invention that is otherwise good. But that itself creates practical mischief as the validity of the patent may turn on fine points of construction of the disclosure. For example, in Miller the question was whether a “steady” light meant steady enough to provide practical illumination for a cyclists, or 50 Hertz, as had previously, but wrongly, been thought necessary for that purpose, p.325. The validity of the patent turned on this point, though word “steady” did not appear at all in the claims.
GlaxoSmithKline / rosiglitazone (NOC) 2011 FC 239 illustrates the problems with the false promise doctrine. The argument rested on the standard Canadian premise that validity turns on “utility as promised in the patent” [97]. The crucial question was whether the patent promised that rosiglitazone “will be useful,” or simply that it “was expected to be useful” [107]. As Hughes J rightly remarked, this debate is “splitting hairs” [107]. Yet the hair splitting was important; at the relevant date the patentee had established only that the compound would be potentially useful. If it had promised more, then the patent would be invalid on the false promise doctrine, even though potential utility, in Hughes J’s holding, was otherwise enough to satisfy the utility requirement. It is disturbing that patent validity should turn on this technical distinction. The real question, surely, should be whether the degree of utility established was sufficient to warrant a patent. This question is difficult enough without adding a further subjective inquiry as to what the patentee promised in the specification.
In his decision, Hughes J avoided the application of the doctrine by the usual method of interpreting the promise of the patent modestly. He held that “improved” meant better than nothing (a control), rather than better that prior art compounds; and that the promise was only that the compound has potential for use in treating diabetes, not that it would treat diabetes [94c]. In the same vein, but of more general interest, he held [94] that the word “surprising” was “self-serving and of little value one way or another in determining what is the real nature of the invention.” In effect, it is mere puffery. This is an important practical holding, given the routine use of this term, though it reinforces the broader point that the promise of the patent will not be construed strictly.
It is also significant that it is construction of the disclosure, not the claims, that are at issue. The role of the claims is to define the scope of the monopoly, and it is reasonable that its construction may be contentious, though even there the courts have cautioned against “meticulous verbal analysis” Catnic [1982] RPC 183, 243. It is perhaps not so reasonable to import the same degree of scrutiny into the construction of the specification itself; the role of the disclosure is to inform the reader about the invention, not to define the scope of the monopoly, so fine distinctions as to the meaning of particular words should not be crucial. As Hughes J pointed out in GlaxoSmithKline / rosiglitazone (NOC) 2011 FC 239 [83], “[t]here has been considerable jurisprudence as to reading a claim, which is part of the overall specification of a patent, but less jurisprudence as to how to read the description; particularly the “promise” of a patent.” Rather than developing a new body of jurisprudence on this point, perhaps it would be better to abandon the false promise doctrine entirely.
1,328,452 / rosiglitazone / AVANDIA
In Canadian law it is standard to say that “[w]here the specification does not promise a specific result, no particular level of utility is required; a ‘mere scintilla’ of utility will suffice. However, where the specification sets out an explicit "promise", utility will be measured against that promise” Eli Lilly Canada / olanzapine 2010 FCA 197 [76]. The notion that the utility is measured by the promise of the patent is now peculiar to Canadian law. It came to us from English law, but it is no longer a consideration in that country: Pharmacia Corp. v Merck & Co. Inc. [2001] EWCA Civ 1610 [54]. While it was raised most commonly in the utility context, it was based on the provision of the earlier Act that a patent might be revoked if “the patent was obtained on a false suggestion or representation” (see Raleigh Cycle v Miller [1948] All ER 308 (HL), 324). Both that provision as a ground of revocation, and the requirement of utility as such, were abolished by the Patents Act, 1977. The doctrine seems never to have been part of US law.
The result of the strict application of the doctrine may be that an inventor with a perfectly good invention will find its patent invalid because the invention’s virtues were exaggerated in the disclosure. This result is rare, as the courts are generally unwilling to read the promise so broadly as to invalidate a patent for an invention that is otherwise good. But that itself creates practical mischief as the validity of the patent may turn on fine points of construction of the disclosure. For example, in Miller the question was whether a “steady” light meant steady enough to provide practical illumination for a cyclists, or 50 Hertz, as had previously, but wrongly, been thought necessary for that purpose, p.325. The validity of the patent turned on this point, though word “steady” did not appear at all in the claims.
GlaxoSmithKline / rosiglitazone (NOC) 2011 FC 239 illustrates the problems with the false promise doctrine. The argument rested on the standard Canadian premise that validity turns on “utility as promised in the patent” [97]. The crucial question was whether the patent promised that rosiglitazone “will be useful,” or simply that it “was expected to be useful” [107]. As Hughes J rightly remarked, this debate is “splitting hairs” [107]. Yet the hair splitting was important; at the relevant date the patentee had established only that the compound would be potentially useful. If it had promised more, then the patent would be invalid on the false promise doctrine, even though potential utility, in Hughes J’s holding, was otherwise enough to satisfy the utility requirement. It is disturbing that patent validity should turn on this technical distinction. The real question, surely, should be whether the degree of utility established was sufficient to warrant a patent. This question is difficult enough without adding a further subjective inquiry as to what the patentee promised in the specification.
In his decision, Hughes J avoided the application of the doctrine by the usual method of interpreting the promise of the patent modestly. He held that “improved” meant better than nothing (a control), rather than better that prior art compounds; and that the promise was only that the compound has potential for use in treating diabetes, not that it would treat diabetes [94c]. In the same vein, but of more general interest, he held [94] that the word “surprising” was “self-serving and of little value one way or another in determining what is the real nature of the invention.” In effect, it is mere puffery. This is an important practical holding, given the routine use of this term, though it reinforces the broader point that the promise of the patent will not be construed strictly.
It is also significant that it is construction of the disclosure, not the claims, that are at issue. The role of the claims is to define the scope of the monopoly, and it is reasonable that its construction may be contentious, though even there the courts have cautioned against “meticulous verbal analysis” Catnic [1982] RPC 183, 243. It is perhaps not so reasonable to import the same degree of scrutiny into the construction of the specification itself; the role of the disclosure is to inform the reader about the invention, not to define the scope of the monopoly, so fine distinctions as to the meaning of particular words should not be crucial. As Hughes J pointed out in GlaxoSmithKline / rosiglitazone (NOC) 2011 FC 239 [83], “[t]here has been considerable jurisprudence as to reading a claim, which is part of the overall specification of a patent, but less jurisprudence as to how to read the description; particularly the “promise” of a patent.” Rather than developing a new body of jurisprudence on this point, perhaps it would be better to abandon the false promise doctrine entirely.
Friday, March 11, 2011
The Vanishing Distinction between Sound Prediction and Demonstrated Utility
GlaxoSmithKline Inc. v. Pharmascience Inc. / rosiglitazone (NOC) 2011 FC 239, Hughes J
The distinction between demonstrated utility and utility based on sound prediction is important, because in the latter case, but not the former, the data supporting utility must be disclosed in the patent itself. I suggested in an earlier post that the distinction is beginning to collapse, as the threshold for what must be disclosed is lowered: in Pfizer / sildenafil (NOC) 2010 FCA 242 the Court of Appeal indicated that it is enough to advert to the data, rather than specifically desribing it. In GlaxoSmithKline / rosiglitazone (NOC) we see the distinction collapsing along a different dimension, as Hughes J held that showing that rosiglitazone was of “potential use” in the treatment of hypoglycaemia was sufficient to establish demonstrated utility [95], [98], [117]. But how different is demonstrating “potential use” from demonstrating a “sound prediction” of use? If we take seriously the distinction between these concepts, then it follows that an inventor who establishes a sound prediction of potential use has therefore satisfied the utility requirement. This seems unreasonable.
As noted, the practical significance of the distinction is whether the data supporting utility must be disclosed in the patent itself. In GlaxoSmithKline / rosiglitazone (NOC), data supporting the utility for rosiglitazone was not disclosed in the patent, so it would have been invalid had utility been based on sound prediction. It is difficult to avoid the conclusion that Hughes J held that “potential use” amounted to demonstrated utility in order to avoid having to hold the patent invalid on this basis. Such contortions do the law no service. In my earlier post I argued that there is no basis in the Act or in policy for drawing this distinction. We now see that it is collapsing on two doctrinal fronts. The key problem is that view that the data supporting utility must be disclosed in the patent itself. It is time to simply recognize that this doctrine, though recent, is wrong. It must be explicitly rejected before the technical distinctions it has engendered become further entrenched in the law.
The distinction between demonstrated utility and utility based on sound prediction is important, because in the latter case, but not the former, the data supporting utility must be disclosed in the patent itself. I suggested in an earlier post that the distinction is beginning to collapse, as the threshold for what must be disclosed is lowered: in Pfizer / sildenafil (NOC) 2010 FCA 242 the Court of Appeal indicated that it is enough to advert to the data, rather than specifically desribing it. In GlaxoSmithKline / rosiglitazone (NOC) we see the distinction collapsing along a different dimension, as Hughes J held that showing that rosiglitazone was of “potential use” in the treatment of hypoglycaemia was sufficient to establish demonstrated utility [95], [98], [117]. But how different is demonstrating “potential use” from demonstrating a “sound prediction” of use? If we take seriously the distinction between these concepts, then it follows that an inventor who establishes a sound prediction of potential use has therefore satisfied the utility requirement. This seems unreasonable.
As noted, the practical significance of the distinction is whether the data supporting utility must be disclosed in the patent itself. In GlaxoSmithKline / rosiglitazone (NOC), data supporting the utility for rosiglitazone was not disclosed in the patent, so it would have been invalid had utility been based on sound prediction. It is difficult to avoid the conclusion that Hughes J held that “potential use” amounted to demonstrated utility in order to avoid having to hold the patent invalid on this basis. Such contortions do the law no service. In my earlier post I argued that there is no basis in the Act or in policy for drawing this distinction. We now see that it is collapsing on two doctrinal fronts. The key problem is that view that the data supporting utility must be disclosed in the patent itself. It is time to simply recognize that this doctrine, though recent, is wrong. It must be explicitly rejected before the technical distinctions it has engendered become further entrenched in the law.
Thursday, March 10, 2011
Primary Screen Establishes Demonstrated Utility
GlaxoSmithKline Inc. v. Pharmascience Inc. / rosiglitazone (NOC) 2011 FC 239 Hughes J
The question of exactly how much utility is necessary to satisfy the utility requirement is a crucial, particularly for pharmaceutical patents. It is well-established that “an inventor is not required to meet regulatory testing standards in order to demonstrate utility”: Pfizer v. Novopharm / sildenafil (NOC) 2010 FCA 242 [97]. It is also reasonably clear that human testing is not required to established utility based on sound prediction: Merck / lovastatin 2010 FC 1265 [510]. In GlaxoSmithKline / rosiglitazone (NOC) Hughes J went further and held [117] that the “primary screen,” described [115] as “the first test that you would, do to look for efficacy [and] the basis on which our department would progress or wish to progress a compound further,” is sufficient to establish demonstrated utility.
The claim in question was to rosiglitazone, an anti-diabetic drug in the thiazolidinedione class of drugs. The primary screen in question was a single murine oral glucose tolerance test, run at only a single dosage level [99-101]. The holding was expressly not based on sound prediction. GSK had not run a sound prediction argument at all [77]; this may have been because the test results on rosiglitazone were not disclosed in the patent itself, as they had reached the patent department too late to be included [32j], and so GSK anticipated trouble with requirement that the factual basis for sound prediction be disclosed. (A similar primary screen on a different compound was the only test data in the patent itself.)
A caveat is that this holding was premised on the particular promise of the patent, which was interpreted [98] as promising only that the claimed compounds were better than nothing (ie a control), and that they were of potential use in treating hypoglycaemia. A patent which promised more might require a correspondingly greater demonstration of utility. (I will discuss problems raised by the focus on the promise of the patent in a subsequent post.) Also, in principle it might be that a different primary screen, or different results, would not satisfy the requirement. Nonetheless, this decision does establish a low threshold for demonstrated utility in pharmaceutical patents. This is consistent with the “scintilla of utility” standard that is generally applicable when the patent itself does not establish a higher threshold: Pfizer v. Novopharm / sildenafil (NOC) 2010 FCA 242 [94].
The question of exactly how much utility is necessary to satisfy the utility requirement is a crucial, particularly for pharmaceutical patents. It is well-established that “an inventor is not required to meet regulatory testing standards in order to demonstrate utility”: Pfizer v. Novopharm / sildenafil (NOC) 2010 FCA 242 [97]. It is also reasonably clear that human testing is not required to established utility based on sound prediction: Merck / lovastatin 2010 FC 1265 [510]. In GlaxoSmithKline / rosiglitazone (NOC) Hughes J went further and held [117] that the “primary screen,” described [115] as “the first test that you would, do to look for efficacy [and] the basis on which our department would progress or wish to progress a compound further,” is sufficient to establish demonstrated utility.
The claim in question was to rosiglitazone, an anti-diabetic drug in the thiazolidinedione class of drugs. The primary screen in question was a single murine oral glucose tolerance test, run at only a single dosage level [99-101]. The holding was expressly not based on sound prediction. GSK had not run a sound prediction argument at all [77]; this may have been because the test results on rosiglitazone were not disclosed in the patent itself, as they had reached the patent department too late to be included [32j], and so GSK anticipated trouble with requirement that the factual basis for sound prediction be disclosed. (A similar primary screen on a different compound was the only test data in the patent itself.)
A caveat is that this holding was premised on the particular promise of the patent, which was interpreted [98] as promising only that the claimed compounds were better than nothing (ie a control), and that they were of potential use in treating hypoglycaemia. A patent which promised more might require a correspondingly greater demonstration of utility. (I will discuss problems raised by the focus on the promise of the patent in a subsequent post.) Also, in principle it might be that a different primary screen, or different results, would not satisfy the requirement. Nonetheless, this decision does establish a low threshold for demonstrated utility in pharmaceutical patents. This is consistent with the “scintilla of utility” standard that is generally applicable when the patent itself does not establish a higher threshold: Pfizer v. Novopharm / sildenafil (NOC) 2010 FCA 242 [94].
Wednesday, March 9, 2011
Abuse of Process: The Door Opens
Apotex Inc. v. Pfizer Ireland Pharmaceuticals / sildenafil 2011 FCA 77 Sexton JA: Layden-Stevenson, Stratas JJA var’g 2010 FC 968, Hughes J
Pfizer Ireland / sildenafil 2011 FCA 77 is an important decision on abuse of process and other pleadings related to the fact that PM(NOC) proceedings and an infringement action are entirely distinct proceedings that nonetheless often raise the same issues between the same parties. The prior FCA case law had undoubtedly “taken a dim view of attempts to prevent relitigation of issues decided in NOC proceedings in subsequent actions” [12]. In Pfizer Ireland / sildenafil 2011 FCA 77 the FCA affirmed that a defendant is in no way estopped from relitigating any cause of action, whether by way of res judicata, issue estoppel, abuse of process or any other doctrine [19]. Indeed, on the pleadings in issue the FCA varied the decision of Hughes J only by striking some paragraphs that he had allowed to stand. This is not surprising.
What is striking and significant is that the FCA went to considerable lengths to hold, after an extensive review of the case-law, that issue estoppel and / or abuse of process may bar relitigation of subsidiary factual and legal issues [24]. The Court specified that this may be particularly “where the evidentiary record at trial was identical to that of the NOC proceeding,” [23] but the Court did not set out any strict rules in this regard, and it noted that more broadly “issue estoppel generally precludes parties from raising arguments or issues that could have been raised at the original hearing,” [25, emphasis added]. The FCA left to the trial judge the question of whether a discretionary bar should apply in respect of a particular issue or factual determination [29], cautioning one the one hand that “courts should be cognisant of the summary nature of NOC proceedings and the fact that no discoveries or live evidence are permissible,” [25], but also that “issue estoppel and abuse of process exist primarily as pragmatic rules intended to promote judicial economy and efficiency. Those who act in a way such that pragmatism, judicial economy and efficiency are adversely affected, may find that the judge exercises his or her discretion in order to prevent such conduct” [27].
The Court’s reasoning is persuasive on its face, both in the distinction between cause of action estoppel and issue estoppel, and in the policy point regarding judicial economy. Nonetheless, there are clearly problems with this approach. The Court “acknowledge[d] a risk that parties may be tempted to make submissions concerning issue estoppel and abuse of process witness by witness, document by document, thereby prolonging proceedings,” but felt that this risk could be controlled by the discretion of the trial judge [27]. Moreover, even if it true that the trial judge can deal more or less adequately with these concerns, it strikes me that this is a second-best approach as compared with a system in which the NOC proceeding, or its equivalent – namely an interlocutory injunction – was actually part of the same proceeding as the infringement action. The problem of duplicative proceedings is just one of the problems that stems, fundamentally, from the distinct nature of NOC proceedings. Of course, that is a problem which cannot be addressed by the FCA.
Another interesting point is that there may be a split in the FCA on this issue, or at least a shift in views, between Pfizer Ireland / sildenafil and Janssen-Ortho v Apotex / levofloxacin (NOC) 2009 FCA 212. In litigation between Janssen-Ortho and Novopharm, Hughes J had found the levoflaxin patent to be valid and infringed, and this was affirmed by the FCA: 2006 FC 1234 affm’d 2007 FCA 217. In subsequent NOC proceedings between Janssen-Ortho and Apotex, Shore J referred repeatedly to the decision of Hughes J in the Novopharm litigation, and, while he did (in my view) carry out an independent review of the evidence, he relied on abuse principles to say "[t]his Court . . . would require better evidence and more appropriate legal argument . . . to come to a different result” 2008 FC 744 [214]. The majority of the FCA in Janssen-Ortho v Apotex / levofloxacin (NOC) remitted the matter back to Shore J to assess the evidence before him “independently of any findings made by Hughes J. in the Novopharm trial.” [80] (See here regarding the eventual fallout.) In contrast, the Pfizer Ireland / sildenafil decision expressly allows the trial judge to consider whether the evidence in the subsequent proceeding is different from that in the prior action. It is perfectly clear that the principles elaborated on in the Pfizer Ireland / sildenafil decision apply equally whether it involves a prior NOC and a subsequent infringement action, as in Pfizer Ireland / sildenafil, or a prior infringement action and a subsequent NOC, as in Janssen-Ortho / levofloxacin (NOC). The decisions cannot be reconciled on this basis. Indeed, generally the weight given to a prior infringement action should be greater, as the proceedings are full rather than summary.
A more plausible distinction is that the levoflaxin litigation involved different parties – Novopharm in the infringement action and Apotex in the NOC proceeding – while the parties to the sildenafil litigation were the same. However, in Pfizer Ireland / sildenafil the FCA noted that one of the rationales for abuse of process is to “promote the integrity of the justice system [and] prevent inconsistent findings” [24]. Surely the system is embarrassed by inconsistent findings based on the same evidence and arguments, whether the parties are the same or not. Similarly, at some point judicial resources are wasted in repeatedly litigating the validity of the same patent, whether or not the parties are the same. If these really are the justifications for the abuse doctrine, Shore J’s requirement that the new parties raise different arguments or different evidence is not so unreasonable.
Layden-Stevenson JA was the only judge in common between the two panels, and she dissented in the levoflaxin decision. However, she concurred on the abuse of process point [81], and dissented only because she felt that this error did not taint his own assessment of the issues. Thus, this is not a case where we can say that there is an obvious split in the Court in the sense that Layden-Stevenson J was in dissent on a point of principle in levoflaxin. Nonetheless, the principled distinction between the cases is not obvious to me. I would be interested to hear how Layden-Stevenson J would reconcile these cases.
Pfizer Ireland / sildenafil 2011 FCA 77 is an important decision on abuse of process and other pleadings related to the fact that PM(NOC) proceedings and an infringement action are entirely distinct proceedings that nonetheless often raise the same issues between the same parties. The prior FCA case law had undoubtedly “taken a dim view of attempts to prevent relitigation of issues decided in NOC proceedings in subsequent actions” [12]. In Pfizer Ireland / sildenafil 2011 FCA 77 the FCA affirmed that a defendant is in no way estopped from relitigating any cause of action, whether by way of res judicata, issue estoppel, abuse of process or any other doctrine [19]. Indeed, on the pleadings in issue the FCA varied the decision of Hughes J only by striking some paragraphs that he had allowed to stand. This is not surprising.
What is striking and significant is that the FCA went to considerable lengths to hold, after an extensive review of the case-law, that issue estoppel and / or abuse of process may bar relitigation of subsidiary factual and legal issues [24]. The Court specified that this may be particularly “where the evidentiary record at trial was identical to that of the NOC proceeding,” [23] but the Court did not set out any strict rules in this regard, and it noted that more broadly “issue estoppel generally precludes parties from raising arguments or issues that could have been raised at the original hearing,” [25, emphasis added]. The FCA left to the trial judge the question of whether a discretionary bar should apply in respect of a particular issue or factual determination [29], cautioning one the one hand that “courts should be cognisant of the summary nature of NOC proceedings and the fact that no discoveries or live evidence are permissible,” [25], but also that “issue estoppel and abuse of process exist primarily as pragmatic rules intended to promote judicial economy and efficiency. Those who act in a way such that pragmatism, judicial economy and efficiency are adversely affected, may find that the judge exercises his or her discretion in order to prevent such conduct” [27].
The Court’s reasoning is persuasive on its face, both in the distinction between cause of action estoppel and issue estoppel, and in the policy point regarding judicial economy. Nonetheless, there are clearly problems with this approach. The Court “acknowledge[d] a risk that parties may be tempted to make submissions concerning issue estoppel and abuse of process witness by witness, document by document, thereby prolonging proceedings,” but felt that this risk could be controlled by the discretion of the trial judge [27]. Moreover, even if it true that the trial judge can deal more or less adequately with these concerns, it strikes me that this is a second-best approach as compared with a system in which the NOC proceeding, or its equivalent – namely an interlocutory injunction – was actually part of the same proceeding as the infringement action. The problem of duplicative proceedings is just one of the problems that stems, fundamentally, from the distinct nature of NOC proceedings. Of course, that is a problem which cannot be addressed by the FCA.
Another interesting point is that there may be a split in the FCA on this issue, or at least a shift in views, between Pfizer Ireland / sildenafil and Janssen-Ortho v Apotex / levofloxacin (NOC) 2009 FCA 212. In litigation between Janssen-Ortho and Novopharm, Hughes J had found the levoflaxin patent to be valid and infringed, and this was affirmed by the FCA: 2006 FC 1234 affm’d 2007 FCA 217. In subsequent NOC proceedings between Janssen-Ortho and Apotex, Shore J referred repeatedly to the decision of Hughes J in the Novopharm litigation, and, while he did (in my view) carry out an independent review of the evidence, he relied on abuse principles to say "[t]his Court . . . would require better evidence and more appropriate legal argument . . . to come to a different result” 2008 FC 744 [214]. The majority of the FCA in Janssen-Ortho v Apotex / levofloxacin (NOC) remitted the matter back to Shore J to assess the evidence before him “independently of any findings made by Hughes J. in the Novopharm trial.” [80] (See here regarding the eventual fallout.) In contrast, the Pfizer Ireland / sildenafil decision expressly allows the trial judge to consider whether the evidence in the subsequent proceeding is different from that in the prior action. It is perfectly clear that the principles elaborated on in the Pfizer Ireland / sildenafil decision apply equally whether it involves a prior NOC and a subsequent infringement action, as in Pfizer Ireland / sildenafil, or a prior infringement action and a subsequent NOC, as in Janssen-Ortho / levofloxacin (NOC). The decisions cannot be reconciled on this basis. Indeed, generally the weight given to a prior infringement action should be greater, as the proceedings are full rather than summary.
A more plausible distinction is that the levoflaxin litigation involved different parties – Novopharm in the infringement action and Apotex in the NOC proceeding – while the parties to the sildenafil litigation were the same. However, in Pfizer Ireland / sildenafil the FCA noted that one of the rationales for abuse of process is to “promote the integrity of the justice system [and] prevent inconsistent findings” [24]. Surely the system is embarrassed by inconsistent findings based on the same evidence and arguments, whether the parties are the same or not. Similarly, at some point judicial resources are wasted in repeatedly litigating the validity of the same patent, whether or not the parties are the same. If these really are the justifications for the abuse doctrine, Shore J’s requirement that the new parties raise different arguments or different evidence is not so unreasonable.
Layden-Stevenson JA was the only judge in common between the two panels, and she dissented in the levoflaxin decision. However, she concurred on the abuse of process point [81], and dissented only because she felt that this error did not taint his own assessment of the issues. Thus, this is not a case where we can say that there is an obvious split in the Court in the sense that Layden-Stevenson J was in dissent on a point of principle in levoflaxin. Nonetheless, the principled distinction between the cases is not obvious to me. I would be interested to hear how Layden-Stevenson J would reconcile these cases.
Tuesday, March 8, 2011
Accounting and Injunction Granted: No News is News
Valence Technology, Inc. v Phostech Lithium Inc. 2011 FC 174, Gauthier J
I noted in a recent post that Snider J, in her lovastatin decision 2010 FC 1265, departed from the common practice of allowing a successful patentee an election between an accounting and damages. What was noteworthy about Snider J’s decision was not simply that the patentee was confined to damages, but that Snider J expressed a presumption against allowing an election, on the basis that “[i]t is necessary for a party seeking an equitable remedy, such as profits, to show some basis for the exercise of equity” [617].
This may be contrasted with Gauthier J’s decision in Valence Technology. The defendants had argued that an accounting should be denied because of undue delay and bad faith. Gauthier J dismissed both of these arguments on the facts, and allowed Valence to elect [234]-[238]. What is significant, is that while Gauthier J expressly recognized the discretionary nature of an accounting, she did not place any special burden on a plaintiff seeking an election. While her reasons on this point were brief, the tenor is clearly that the plaintiff will be allowed an election unless there is some particular reason to refuse it. This contrasts with the view expressed by Snider J that an accounting will only be permitted if there is some particular reason to allow it. In this respect, Gauthier J’s decision reaffirms the orthodox practice.
Gauthier J also granted injunctive relief [239]-[240]. This is the invariable practice and would hardly be worth mentioning, save for the question of whether the US Supreme Court decision in eBay Inc v MercExchange, LLC, 547 U.S. 388 (2006) will make its impact felt in Canada. In Valence Technology, the defendant had asked the Court to give it a grace period of approximately two years before giving effect to the injunction, in order to give it time to finish a new factory that was designed to use a non-infringing process. The defendant in this case is relatively sympathetic, in that it was clear on the facts that it had developed the infringing process independently. Of course, independent creation is not a substantive defence, but given that injunctive relief is equitable, in principle independent creation might be a relevant consideration. On the other hand, the patentee in this case, who had developed and was practising the invention itself, is also much more sympathetic than the plaintiff in MercExchange. It is interesting, but not surprising, that we will have to wait for a stronger set of facts before the possibility of refusing final injunctive relief to a successful patentee becomes a real issue.
I noted in a recent post that Snider J, in her lovastatin decision 2010 FC 1265, departed from the common practice of allowing a successful patentee an election between an accounting and damages. What was noteworthy about Snider J’s decision was not simply that the patentee was confined to damages, but that Snider J expressed a presumption against allowing an election, on the basis that “[i]t is necessary for a party seeking an equitable remedy, such as profits, to show some basis for the exercise of equity” [617].
This may be contrasted with Gauthier J’s decision in Valence Technology. The defendants had argued that an accounting should be denied because of undue delay and bad faith. Gauthier J dismissed both of these arguments on the facts, and allowed Valence to elect [234]-[238]. What is significant, is that while Gauthier J expressly recognized the discretionary nature of an accounting, she did not place any special burden on a plaintiff seeking an election. While her reasons on this point were brief, the tenor is clearly that the plaintiff will be allowed an election unless there is some particular reason to refuse it. This contrasts with the view expressed by Snider J that an accounting will only be permitted if there is some particular reason to allow it. In this respect, Gauthier J’s decision reaffirms the orthodox practice.
Gauthier J also granted injunctive relief [239]-[240]. This is the invariable practice and would hardly be worth mentioning, save for the question of whether the US Supreme Court decision in eBay Inc v MercExchange, LLC, 547 U.S. 388 (2006) will make its impact felt in Canada. In Valence Technology, the defendant had asked the Court to give it a grace period of approximately two years before giving effect to the injunction, in order to give it time to finish a new factory that was designed to use a non-infringing process. The defendant in this case is relatively sympathetic, in that it was clear on the facts that it had developed the infringing process independently. Of course, independent creation is not a substantive defence, but given that injunctive relief is equitable, in principle independent creation might be a relevant consideration. On the other hand, the patentee in this case, who had developed and was practising the invention itself, is also much more sympathetic than the plaintiff in MercExchange. It is interesting, but not surprising, that we will have to wait for a stronger set of facts before the possibility of refusing final injunctive relief to a successful patentee becomes a real issue.
Friday, March 4, 2011
Caution Regarding Inequitable Conduct
Valence Technology, Inc. v Phostech Lithium Inc. 2011 FC 174, Gauthier J
Inequitable conduct in patent prosecution poses a thorny problem. On one hand, the practice of willfully misleading the patent office during prosecution in order to obtain a patent that would not have been granted must be discouraged. On the other hand, introducing an intent element into the determination of patent validity introduces uncertainty and expense to the litigation, and the sanction of invalidity is a severe punishment for bad faith in cases where the misinformation is substantively inconsequential. The US has been wrestling with this problem for years, and an en banc decision from the Federal Circuit in Therasense, Inc v Becton, Dickinson and Co is in the offing.
In contrast, Canadian courts have historically been reluctant to invalidate patents on this basis. Recently, however, Hughes J has been seeking to give life to this ground of attack, on the basis of both the older section 53(1) and the new section 73(1)(a). In GD Searle & Co v Novopharm Ltd / celecoxib (NOC) 2007 FC 81, Hughes J held a patent to be invalid on the basis of section 73. While he was reversed, without comment on this point, 2007 FCA 173, Malone JA: Noël, Sexton JJA, his decision was picked up in Mactavish J’s controversial decision in Lundbeck Canada Inc v Ratiopharm Inc / memantine (NOC), 2009 FC 1102. In Ratiopharm Inc. v Pfizer Ltd. / amlodipine besylate, 2009 FC 711, [195]-[204], Hughes J held a patent to be invalid on the basis of section 53(1). This was affirmed on other grounds 2010 FCA 204.
In Valence Technology, Inc v Phostech Lithium Inc 2011 FC 174, Gauthier J indirectly cast doubt on Hughes J’s decision in Ratiopharm v Pfizer respecting section 53. She stated [184] that “[i]t is important to mention that the comments of Justice Roger Hughes in Ratiopharm Inc. v Pfizer Ltd., 2009 FC 711, in respect of subsection 27(3) (as well as those relating to ss. 53(1)) were described by the Federal Court of Appeal as simple ‘obiter’ (2010 FCA 204 at para 2).” She then went on to distinguish Hughes J’s decision on the facts, saying she was satisfied that the case before her was substantially different, “regardless of the intrinsic value of the comments made by Justice Hughes on the law” [200].
It is certainly true that the two decisions can be easily distinguished on the facts. Phostech’s misconduct argument was curious and unique. It argued that Valence’s misconduct lay in copying the wording of Phostech’s subsequent patent application in the divisional to Valence’s prior application. Gauthier J explained that there is simply nothing wrong with this: “there is no copyright on claim language” [211]. That the cases are so readily distinguishable on the facts gives added significance to Gauthier J’s subtle remarks regarding Hughes J’s analysis.
It may be reading too much into a couple of phrases, and we certainly cannot draw any firm conclusions as to Gauthier J’s own views regarding inequitable conduct, but Valence Technology suggests that Gauthier J, for one, will be cautious before following the path that Hughes J seeks to blaze.
Inequitable conduct in patent prosecution poses a thorny problem. On one hand, the practice of willfully misleading the patent office during prosecution in order to obtain a patent that would not have been granted must be discouraged. On the other hand, introducing an intent element into the determination of patent validity introduces uncertainty and expense to the litigation, and the sanction of invalidity is a severe punishment for bad faith in cases where the misinformation is substantively inconsequential. The US has been wrestling with this problem for years, and an en banc decision from the Federal Circuit in Therasense, Inc v Becton, Dickinson and Co is in the offing.
In contrast, Canadian courts have historically been reluctant to invalidate patents on this basis. Recently, however, Hughes J has been seeking to give life to this ground of attack, on the basis of both the older section 53(1) and the new section 73(1)(a). In GD Searle & Co v Novopharm Ltd / celecoxib (NOC) 2007 FC 81, Hughes J held a patent to be invalid on the basis of section 73. While he was reversed, without comment on this point, 2007 FCA 173, Malone JA: Noël, Sexton JJA, his decision was picked up in Mactavish J’s controversial decision in Lundbeck Canada Inc v Ratiopharm Inc / memantine (NOC), 2009 FC 1102. In Ratiopharm Inc. v Pfizer Ltd. / amlodipine besylate, 2009 FC 711, [195]-[204], Hughes J held a patent to be invalid on the basis of section 53(1). This was affirmed on other grounds 2010 FCA 204.
In Valence Technology, Inc v Phostech Lithium Inc 2011 FC 174, Gauthier J indirectly cast doubt on Hughes J’s decision in Ratiopharm v Pfizer respecting section 53. She stated [184] that “[i]t is important to mention that the comments of Justice Roger Hughes in Ratiopharm Inc. v Pfizer Ltd., 2009 FC 711, in respect of subsection 27(3) (as well as those relating to ss. 53(1)) were described by the Federal Court of Appeal as simple ‘obiter’ (2010 FCA 204 at para 2).” She then went on to distinguish Hughes J’s decision on the facts, saying she was satisfied that the case before her was substantially different, “regardless of the intrinsic value of the comments made by Justice Hughes on the law” [200].
It is certainly true that the two decisions can be easily distinguished on the facts. Phostech’s misconduct argument was curious and unique. It argued that Valence’s misconduct lay in copying the wording of Phostech’s subsequent patent application in the divisional to Valence’s prior application. Gauthier J explained that there is simply nothing wrong with this: “there is no copyright on claim language” [211]. That the cases are so readily distinguishable on the facts gives added significance to Gauthier J’s subtle remarks regarding Hughes J’s analysis.
It may be reading too much into a couple of phrases, and we certainly cannot draw any firm conclusions as to Gauthier J’s own views regarding inequitable conduct, but Valence Technology suggests that Gauthier J, for one, will be cautious before following the path that Hughes J seeks to blaze.
Thursday, March 3, 2011
Claim Construction and Validity
Valence Technology, Inc. v. Phostech Lithium Inc. 2011 FC 174, Gauthier J
In Valence Technology Gauthier J remarked at [138] that “the Court must construe the claim without regard to its validity.” I suggest that this is not correct. The Supreme Court has said that “the guiding principle” of purposive construction is that “where the language of the specification, upon a reasonable view of it, can be so read as to afford the inventor protection for that which he has actually in good faith invented, the court, as a rule, will endeavour to give effect to that construction."1 Many other statements of high authority, ancient and modern, are to the same effect: a patent should be approached "with a judicial anxiety to support a really useful invention"2; “such construction ought to be made as will, consistently with the fair import of the language used, make the claim of invention co-extensive with the new discovery of the grantee of the patent”3;“it is the widest purpose consistent with [the patentee’s] teaching which should be used for purposive construction."4
Thus the true rule is that the claims should be construed with a view to validly claiming that which the inventor has invented and disclosed. This follows from the general rule that legal documents, from legislation, to patents, to contracts, are to be of construed in light of the drafter’s purpose. The skilled reader does not just understand the technical terms that are used in the patent; she also understands the technical advance made by the patentee, and she understands that the patentee intends to claim that advance. For a more extended discussion, see my recent article, “The Essential Elements Doctrine in Patent Infringement: Free World and Whirlpool in Light of Kirin-Amgen” 22 IPJ 223 - 251 (2010).
With that said, I have been unfair to Gauthier J in taking her statement out of context. Phostech was arguing for a construction that would have rendered the claim invalid, and in saying that the Court must construe the claim “without regard to its validity,” Gauthier J was saying in effect that the Court should not go out of its way to construe the claim in a manner that will result in invalidity. This is entirely consistent with the authorities cited above.
Footnotes are after the break.
In Valence Technology Gauthier J remarked at [138] that “the Court must construe the claim without regard to its validity.” I suggest that this is not correct. The Supreme Court has said that “the guiding principle” of purposive construction is that “where the language of the specification, upon a reasonable view of it, can be so read as to afford the inventor protection for that which he has actually in good faith invented, the court, as a rule, will endeavour to give effect to that construction."1 Many other statements of high authority, ancient and modern, are to the same effect: a patent should be approached "with a judicial anxiety to support a really useful invention"2; “such construction ought to be made as will, consistently with the fair import of the language used, make the claim of invention co-extensive with the new discovery of the grantee of the patent”3;“it is the widest purpose consistent with [the patentee’s] teaching which should be used for purposive construction."4
Thus the true rule is that the claims should be construed with a view to validly claiming that which the inventor has invented and disclosed. This follows from the general rule that legal documents, from legislation, to patents, to contracts, are to be of construed in light of the drafter’s purpose. The skilled reader does not just understand the technical terms that are used in the patent; she also understands the technical advance made by the patentee, and she understands that the patentee intends to claim that advance. For a more extended discussion, see my recent article, “The Essential Elements Doctrine in Patent Infringement: Free World and Whirlpool in Light of Kirin-Amgen” 22 IPJ 223 - 251 (2010).
With that said, I have been unfair to Gauthier J in taking her statement out of context. Phostech was arguing for a construction that would have rendered the claim invalid, and in saying that the Court must construe the claim “without regard to its validity,” Gauthier J was saying in effect that the Court should not go out of its way to construe the claim in a manner that will result in invalidity. This is entirely consistent with the authorities cited above.
Footnotes are after the break.
Wednesday, March 2, 2011
Federal Court Overrules Supreme Court on Presumption of Validity
Eli Lilly Co. v. Apotex Inc. / cefaclor, 2009 FC 991, 80 C.P.R. (4th) 1, Gauthier J. aff’d 2010 FCA 240
Valence Technology, Inc. v. Phostech Lithium Inc. 2011 FC 174, Gauthier J
In Wellcome / AZT 2002 SCC 77 [44], the Supreme Court held that the appropriate “standard of review” of a granted patent on a question of mixed fact and law, such as validity challenge based on utility, is “reasonableness simpliciter.” This was a importation of administrative law principles into what had previously been dealt with in terms of the statutory presumption of validity set out in ss 43(2). The novelty of the Court’s approach is reflected even in the heading to its discussion, “The Standard of Review,” in contrast to the traditional “presumption of validity.”
In Eli Lilly / cefaclor 2009 FC 991 Gauthier J refused to follow this holding. Gauthier J’s discussion on this point is one of my favourite passages of judicial reasoning from the past year, for two reasons. First, her analysis is crushing, in terms of law and policy. In my view it is quite clear that Gauthier J is right and the Supreme Court is wrong. I have nothing to add to her thorough reasons, which I have reproduced below. Of course, neither strong reasoning in the Federal Court, nor weak reasoning in the Supreme Court, are novel in themselves. What makes Gauthier J’s analysis stand out is her willingness to face the issue directly. The more usual course when a lower court is faced with unpalatable authority is to interpret that authority narrowly, or make factual findings that avoid its application, until the inconvenient doctrine fades away from lack of use. This strategy has the advantage of formally respecting binding authority, but the corresponding disadvantage that the problematic doctrine distorts the law as the courts frame their reasons so as to avoid the doctrine, rather than to illuminate the issues. (As another example, see here.) Of course, Gauthier J’s decision not to follow the Supreme Court’s decision can be criticized as going beyond the role of a trial judge. But, as she pointed out, she did have very recent Supreme Court authority on her side, which the Supreme Court in Wellcome / AZT had not taken into account, and, moreover, the point had not been argued by the parties in Wellcome / AZT. It seems quite likely that the SCC did not realize it was making new law.
Gauthier J’s position reflects the general trend at the Federal Court. In Valence Technology Gauthier J at [179] reaffirmed the analysis she had set out in Eli Lilly / cefaclor. In Novopharm Ltd. v. Eli Lilly & Co. / tomoxetine 2010 FC 915 [29-30], Barnes J expressly followed Gauthier J in declining to follow Wellcome / AZT. While his reasons were less elaborate, they were equally pointed: “It is not entirely clear to me what was meant by Justice Ian Binnie in the discussion in AZT, above, about the administrative standard of review,” and “any deference is owed to the Commissioner in cases like this one, it is completely subsumed by the presumption of validity created by ss. 43(2) of the Patent Act, and is essentially extinguished where any evidence to the contrary is placed before the Court.” As Gauthier J pointed out in Eli Lilly / cefaclor, Snider J had also declined to apply a more deferential standard in 2007 FC 358 and 2009 FC 676, albeit without expressly disagreeing with the Supreme Court. In Eli Lilly Canada Inc. v. Novopharm Ltd. / olanzapine, 2009 FC 1018, 78 C.P.R. (4th) 1 O'Reilly J stated a deferential standard [12], citing Wellcome / AZT, but then apparently applied a balance of probabilities standard [13]. On the other hand, in Blair v Canada (Attorney General), 2010 FC 227, 81 CPR(4th) 403, Mactavish J applied a reasonableness standard, but that was a review of the Commissioner’s decision to refuse a patent, and not a validity challenge to a granted patent. There is a stronger argument for deference in such a case, as the same record is before Court as was before the Commissioner. (However, the point must be considered open, as Mactavish J ultimately held the decision to be unreasonable, so the standard of review did not affect the outcome.)
The important lesson from this decision, I suggest, is one for the Supreme Court. It should stop deciding points of law that are not argued by the parties. This is a simple step that would save the Court, and the law, from this kind of error.
I note that the presumption of validity will be addressed by the US Supreme Court in i4i Ltd. v. Microsoft Corp., 598 F.3d 831 (Fed. Cir. 2010), cert. granted, 131 S.Ct. 647 (Nov. 29, 2010) (No. 10-290). US law has traditionally has a relatively high presumption of validity which requires a showing of invalidity on the basis of “clear and convincing evidence.” That standard is the point on which cert was granted. The current US standard is high relative to the traditional Canadian rule, expressed by Gauthier J, and closer to the standard expressed by the SCC in Wellcome / AZT. My own view is that the current Canadian law, as expressed by Gauthier J, is sound in policy. It will be interesting to see whether the USSC agrees.
The relevant excerpts from cefaclor are after the break.
Valence Technology, Inc. v. Phostech Lithium Inc. 2011 FC 174, Gauthier J
In Wellcome / AZT 2002 SCC 77 [44], the Supreme Court held that the appropriate “standard of review” of a granted patent on a question of mixed fact and law, such as validity challenge based on utility, is “reasonableness simpliciter.” This was a importation of administrative law principles into what had previously been dealt with in terms of the statutory presumption of validity set out in ss 43(2). The novelty of the Court’s approach is reflected even in the heading to its discussion, “The Standard of Review,” in contrast to the traditional “presumption of validity.”
In Eli Lilly / cefaclor 2009 FC 991 Gauthier J refused to follow this holding. Gauthier J’s discussion on this point is one of my favourite passages of judicial reasoning from the past year, for two reasons. First, her analysis is crushing, in terms of law and policy. In my view it is quite clear that Gauthier J is right and the Supreme Court is wrong. I have nothing to add to her thorough reasons, which I have reproduced below. Of course, neither strong reasoning in the Federal Court, nor weak reasoning in the Supreme Court, are novel in themselves. What makes Gauthier J’s analysis stand out is her willingness to face the issue directly. The more usual course when a lower court is faced with unpalatable authority is to interpret that authority narrowly, or make factual findings that avoid its application, until the inconvenient doctrine fades away from lack of use. This strategy has the advantage of formally respecting binding authority, but the corresponding disadvantage that the problematic doctrine distorts the law as the courts frame their reasons so as to avoid the doctrine, rather than to illuminate the issues. (As another example, see here.) Of course, Gauthier J’s decision not to follow the Supreme Court’s decision can be criticized as going beyond the role of a trial judge. But, as she pointed out, she did have very recent Supreme Court authority on her side, which the Supreme Court in Wellcome / AZT had not taken into account, and, moreover, the point had not been argued by the parties in Wellcome / AZT. It seems quite likely that the SCC did not realize it was making new law.
Gauthier J’s position reflects the general trend at the Federal Court. In Valence Technology Gauthier J at [179] reaffirmed the analysis she had set out in Eli Lilly / cefaclor. In Novopharm Ltd. v. Eli Lilly & Co. / tomoxetine 2010 FC 915 [29-30], Barnes J expressly followed Gauthier J in declining to follow Wellcome / AZT. While his reasons were less elaborate, they were equally pointed: “It is not entirely clear to me what was meant by Justice Ian Binnie in the discussion in AZT, above, about the administrative standard of review,” and “any deference is owed to the Commissioner in cases like this one, it is completely subsumed by the presumption of validity created by ss. 43(2) of the Patent Act, and is essentially extinguished where any evidence to the contrary is placed before the Court.” As Gauthier J pointed out in Eli Lilly / cefaclor, Snider J had also declined to apply a more deferential standard in 2007 FC 358 and 2009 FC 676, albeit without expressly disagreeing with the Supreme Court. In Eli Lilly Canada Inc. v. Novopharm Ltd. / olanzapine, 2009 FC 1018, 78 C.P.R. (4th) 1 O'Reilly J stated a deferential standard [12], citing Wellcome / AZT, but then apparently applied a balance of probabilities standard [13]. On the other hand, in Blair v Canada (Attorney General), 2010 FC 227, 81 CPR(4th) 403, Mactavish J applied a reasonableness standard, but that was a review of the Commissioner’s decision to refuse a patent, and not a validity challenge to a granted patent. There is a stronger argument for deference in such a case, as the same record is before Court as was before the Commissioner. (However, the point must be considered open, as Mactavish J ultimately held the decision to be unreasonable, so the standard of review did not affect the outcome.)
The important lesson from this decision, I suggest, is one for the Supreme Court. It should stop deciding points of law that are not argued by the parties. This is a simple step that would save the Court, and the law, from this kind of error.
I note that the presumption of validity will be addressed by the US Supreme Court in i4i Ltd. v. Microsoft Corp., 598 F.3d 831 (Fed. Cir. 2010), cert. granted, 131 S.Ct. 647 (Nov. 29, 2010) (No. 10-290). US law has traditionally has a relatively high presumption of validity which requires a showing of invalidity on the basis of “clear and convincing evidence.” That standard is the point on which cert was granted. The current US standard is high relative to the traditional Canadian rule, expressed by Gauthier J, and closer to the standard expressed by the SCC in Wellcome / AZT. My own view is that the current Canadian law, as expressed by Gauthier J, is sound in policy. It will be interesting to see whether the USSC agrees.
The relevant excerpts from cefaclor are after the break.
Tuesday, March 1, 2011
Valence Tech v Phostech Lithium Overview
Valence Technology, Inc. v. Phostech Lithium Inc. 2011 FC 174, Gauthier J
Gauthier J’s decision in Valence Technology, Inc. v. Phostech Lithium turned almost entirely on the facts, though some interesting legal points were raised tangentially. This post provides an overview. I will discuss the legal issues in separate posts.
The invention in question related to a process for making lithium mixed metal cathodes for lithium ion batteries. It was known in the prior art that LiFePO4 is a good cathode material [24], but it had not been commercially adopted, apparently because of the cost of production. Both the plaintiff, Valence, and the defendant Phostech, had developed new processes for making LiFePO4 cathodes at about the same time. The key to Valence’s process was the use of carbothermal reduction (CTR). Valence held three process patents on its technology: 2,395,115 and 2,466,366, which was a divisional of ‘115, and the subsequent 2,483,918 patent which was intended to claim an extension of the technology to a wider set of compounds.
Challenges to the validity of the ‘115 patent were abandoned in oral argument [180], and Gauthier J held that Phostech’s process infringed the ‘115 patent [178]. While Phostech did raise some claim construction arguments, these were not particularly strong, in my view. While Gauthier J noted [139] that "as a whole the claims are not particularly well written,” she had no real difficulty in coming to a conclusion which favoured Valence. The difficult question was whether the Phostech process used carbothermal reduction or some other reductive process. This was a difficult question on the facts. While there were disputes in discovery, the real difficulty was more fundamental. It was established that the compounds used in the Phostech process could support CTR under the right conditions [158], but it was difficult, as a matter of science, to determine whether the conditions inside the closed industrial kiln used by Phostech would actually result in carbothermal reduction. A variety of tests were conducted by both parties, and ultimately Gauthier J held that Valence had established on the balance of probabilities that the Phostech process used CTR [166]. While Valence’s expert witness on this point was particularly good, and Phostech’s expert was particularly weak [49, 166], Gauthier J carefully reviewed the evidence and from the decision itself there is no reason to believe that the inexperience of Phostech’s expert influenced the outcome. Gauthier J did note that "what surprises me most is the fact that Phostech’s expert did not perform tests that more closely mirror the actual timing and temperatures used in its P1 Process” [168]. Read in context this does not amount to shifting the burden to the defendant, as the plaintiffs had conducted enough tests to support Gauthier J’s conclusion. It does make the basic point that once the plaintiff has advanced enough evidence to support its case on the balance of probabilities, the defendant needs to respond with solid evidence in order to tilt the balance back in the other direction.
Gauthier J made no determination as to infringement of the ‘366 and ‘918 patents, but she did hold that the ‘918 patent was invalid as being anticipated by Canadian Patent Application 2,423,129, which had been published prior to the ‘918 priority date (but after the ‘115 priority date). She had no difficulty in coming to this conclusion. The root of the problem was that the ‘918 patent claimed too broadly. The advance in the ‘918 patent was an extension of the CTR technology to a wider set of compounds, but as drafted it included all the compounds described in the ‘115 patent, and also LiFePO4 which was described in the an example of the ‘129 application [230], and construction was not disputed [151]. (Obviousness of the ‘918 patent over the ‘129 application was not discussed, presumably because the application, while prior art, was not part of the cgk [149].) As all claims of the ‘918 patent were invalid, the patent as a whole was declared to be void.
An insufficiency attack on the ‘366 patent was readily dismissed on the facts [182ff].
In the result, Gauthier J held that ‘115 patent was infringed, and that the ‘918 patent was void.
Gauthier J’s decision in Valence Technology, Inc. v. Phostech Lithium turned almost entirely on the facts, though some interesting legal points were raised tangentially. This post provides an overview. I will discuss the legal issues in separate posts.
The invention in question related to a process for making lithium mixed metal cathodes for lithium ion batteries. It was known in the prior art that LiFePO4 is a good cathode material [24], but it had not been commercially adopted, apparently because of the cost of production. Both the plaintiff, Valence, and the defendant Phostech, had developed new processes for making LiFePO4 cathodes at about the same time. The key to Valence’s process was the use of carbothermal reduction (CTR). Valence held three process patents on its technology: 2,395,115 and 2,466,366, which was a divisional of ‘115, and the subsequent 2,483,918 patent which was intended to claim an extension of the technology to a wider set of compounds.
Challenges to the validity of the ‘115 patent were abandoned in oral argument [180], and Gauthier J held that Phostech’s process infringed the ‘115 patent [178]. While Phostech did raise some claim construction arguments, these were not particularly strong, in my view. While Gauthier J noted [139] that "as a whole the claims are not particularly well written,” she had no real difficulty in coming to a conclusion which favoured Valence. The difficult question was whether the Phostech process used carbothermal reduction or some other reductive process. This was a difficult question on the facts. While there were disputes in discovery, the real difficulty was more fundamental. It was established that the compounds used in the Phostech process could support CTR under the right conditions [158], but it was difficult, as a matter of science, to determine whether the conditions inside the closed industrial kiln used by Phostech would actually result in carbothermal reduction. A variety of tests were conducted by both parties, and ultimately Gauthier J held that Valence had established on the balance of probabilities that the Phostech process used CTR [166]. While Valence’s expert witness on this point was particularly good, and Phostech’s expert was particularly weak [49, 166], Gauthier J carefully reviewed the evidence and from the decision itself there is no reason to believe that the inexperience of Phostech’s expert influenced the outcome. Gauthier J did note that "what surprises me most is the fact that Phostech’s expert did not perform tests that more closely mirror the actual timing and temperatures used in its P1 Process” [168]. Read in context this does not amount to shifting the burden to the defendant, as the plaintiffs had conducted enough tests to support Gauthier J’s conclusion. It does make the basic point that once the plaintiff has advanced enough evidence to support its case on the balance of probabilities, the defendant needs to respond with solid evidence in order to tilt the balance back in the other direction.
Gauthier J made no determination as to infringement of the ‘366 and ‘918 patents, but she did hold that the ‘918 patent was invalid as being anticipated by Canadian Patent Application 2,423,129, which had been published prior to the ‘918 priority date (but after the ‘115 priority date). She had no difficulty in coming to this conclusion. The root of the problem was that the ‘918 patent claimed too broadly. The advance in the ‘918 patent was an extension of the CTR technology to a wider set of compounds, but as drafted it included all the compounds described in the ‘115 patent, and also LiFePO4 which was described in the an example of the ‘129 application [230], and construction was not disputed [151]. (Obviousness of the ‘918 patent over the ‘129 application was not discussed, presumably because the application, while prior art, was not part of the cgk [149].) As all claims of the ‘918 patent were invalid, the patent as a whole was declared to be void.
An insufficiency attack on the ‘366 patent was readily dismissed on the facts [182ff].
In the result, Gauthier J held that ‘115 patent was infringed, and that the ‘918 patent was void.
Thursday, February 24, 2011
Separation of NOC Proceedings from Infringement Action Results in Unfairness in Section 8 Damages
Apotex v Janssen-Ortho Inc / levofloxacin (NOC) 2011 FCA 57 (Stratas JA: Sexton, Layden-Stevenson JJA) varying 2010 FC 711 (Hughes J)
In the levoflaxin dispute, Janssen-Ortho faces the possibility of liability for section 8 damages in respect of a patent that has been found valid in both an infringement action and in the course of the NOC proceedings in which the liability may arise. While the particular facts of the levoflaxin dispute are unusual, the root cause of this paradox is the separation of the stay under the PM(NOC) Regulations and the infringement action itself.
A prohibition order consequent on PM(NOC) proceeding is analogous to an interlocutory injunction, and section 8 is analogous to the undertaking required of the party who obtains such an injunction: 2008 FC 1185 [54]. One important difference is that in contrast to a motion for an interlocutory injunction, the PM(NOC) proceedings are entirely separate from any subsequent infringement action. This means that entitlement to section 8 damages turns on success in the NOC proceedings, not on the success in the infringement action which typically follows. Compared with a true interlocutory injunction, this is unsatisfactory in two ways. On the one hand, if the patentee obtains the prohibition order but the patent is ultimately held to be invalid, the generic will not be able to recover its lost profits from being wrongly excluded from the market (Apotex v Syntex / naproxen (NOC) 2010 FCA 155 affm’g 2009 FC 494), though it would have been able to recover such losses on the undertaking if the PM(NOC) system did not exist, and an interlocutory injunction had been granted to patentee.1 Conversely, if the patentee loses in the NOC proceedings, but prevails in the infringement action, the patentee will nonetheless be liable to the generic under section 8, though it would not have been liable on the undertaking on an interlocutory injunction. This follows both from the general principle that the NOC proceedings and the infringement action are separate, and also on the face of section 8 itself (see esp. subsection 8(3)).
This second scenario has not yet arisen directly, but the levoflaxin litigation raises the same problem in an exaggerated form. In 2004 FC 1631 the patent in issue was found to be invalid for obviousness in NOC proceedings against Novopharm. In the subsequent infringement action against Novopharm, 2006 FC 1234 affm’d 2007 FCA 217, Hughes J found the patent to be valid and infringed. Apotex then applied for an NOC and in 2008 FC 744 Shore J held the patent to be valid and granted a prohibition order. Here is where things took an unusual twist. In the course his decision, Shore J referred repeatedly to the decision of Hughes J in the Novopharm litigation. In 2009 FCA 212 the FCA decided that Shore J had relied excessively on the decision of Hughes J and remitted the matter back to him with instructions to assess the evidence independently. Shore J, evidently piqued at the FCA’s holding, recused himself: 2010 FCA 643 affm’d 2011 FCA 58. The matter was then heard by Hughes J. However, the day after the FCA decision remitting the matter to Shore J, the patent expired, and the day after that the NOC was granted. In 2010 FC 711 Hughes J held the prohibition proceeding to be moot as the NOC had been granted. However, Hughes J did not “dismiss” the action; instead, he “terminated” it for mootness. In 2011 FCA 57 the FCA varied this by holding that the matter was instead “dismissed.” The significance of the terminological distinction is that section 8 provides that the generic is entitled to damages if the application is “dismissed by the court hearing the application.” It was expressly in order to make it clear that Apotex was not entitled to section 8 damages that Hughes J held that the proceeding was “terminated” [34].
In the levoflaxin dispute, Janssen-Ortho faces the possibility of liability for section 8 damages in respect of a patent that has been found valid in both an infringement action and in the course of the NOC proceedings in which the liability may arise. While the particular facts of the levoflaxin dispute are unusual, the root cause of this paradox is the separation of the stay under the PM(NOC) Regulations and the infringement action itself.
A prohibition order consequent on PM(NOC) proceeding is analogous to an interlocutory injunction, and section 8 is analogous to the undertaking required of the party who obtains such an injunction: 2008 FC 1185 [54]. One important difference is that in contrast to a motion for an interlocutory injunction, the PM(NOC) proceedings are entirely separate from any subsequent infringement action. This means that entitlement to section 8 damages turns on success in the NOC proceedings, not on the success in the infringement action which typically follows. Compared with a true interlocutory injunction, this is unsatisfactory in two ways. On the one hand, if the patentee obtains the prohibition order but the patent is ultimately held to be invalid, the generic will not be able to recover its lost profits from being wrongly excluded from the market (Apotex v Syntex / naproxen (NOC) 2010 FCA 155 affm’g 2009 FC 494), though it would have been able to recover such losses on the undertaking if the PM(NOC) system did not exist, and an interlocutory injunction had been granted to patentee.1 Conversely, if the patentee loses in the NOC proceedings, but prevails in the infringement action, the patentee will nonetheless be liable to the generic under section 8, though it would not have been liable on the undertaking on an interlocutory injunction. This follows both from the general principle that the NOC proceedings and the infringement action are separate, and also on the face of section 8 itself (see esp. subsection 8(3)).
This second scenario has not yet arisen directly, but the levoflaxin litigation raises the same problem in an exaggerated form. In 2004 FC 1631 the patent in issue was found to be invalid for obviousness in NOC proceedings against Novopharm. In the subsequent infringement action against Novopharm, 2006 FC 1234 affm’d 2007 FCA 217, Hughes J found the patent to be valid and infringed. Apotex then applied for an NOC and in 2008 FC 744 Shore J held the patent to be valid and granted a prohibition order. Here is where things took an unusual twist. In the course his decision, Shore J referred repeatedly to the decision of Hughes J in the Novopharm litigation. In 2009 FCA 212 the FCA decided that Shore J had relied excessively on the decision of Hughes J and remitted the matter back to him with instructions to assess the evidence independently. Shore J, evidently piqued at the FCA’s holding, recused himself: 2010 FCA 643 affm’d 2011 FCA 58. The matter was then heard by Hughes J. However, the day after the FCA decision remitting the matter to Shore J, the patent expired, and the day after that the NOC was granted. In 2010 FC 711 Hughes J held the prohibition proceeding to be moot as the NOC had been granted. However, Hughes J did not “dismiss” the action; instead, he “terminated” it for mootness. In 2011 FCA 57 the FCA varied this by holding that the matter was instead “dismissed.” The significance of the terminological distinction is that section 8 provides that the generic is entitled to damages if the application is “dismissed by the court hearing the application.” It was expressly in order to make it clear that Apotex was not entitled to section 8 damages that Hughes J held that the proceeding was “terminated” [34].
Monday, February 21, 2011
Maintenance Fees Must be Paid by the Authorized Correspondent: Why?
Unicrop Ltd. v. Canada (Attorney General) 2011 FCA 55 Noël JA: Pelletier, Trudel JJA affm’g 2010 FC 61 Boivin J
In Unicrop the FCA affirmed that maintenance fees must be paid by the authorized correspondent, and that the Commissioner is entitled, and indeed apparently required, to refuse payment from anyone else. It is difficult to argue with this conclusion, given the text of the Patent Rules; the question is why the Rules impose such a stringent requirement.
Unicrop’s representatives on filing were Bereskin & Parr LLP, who also paid the first two annual maintenance fees. The third annual fee was not paid in time and the application was deemed abandoned. Just prior to the one year grace period for reinstatement, the applicant attempted to reinstate the application through their new agent, Furman & Kallio. However, the notice of appointment of agent had not been submitted to CIPO prior to the grace period deadline [FC 27]. CIPO therefore refused to accept the fee payment, on the basis that it had not been submitted by the authorized correspondent. The Appointment of Agent form was submitted to CIPO after the deadline, but CIPO was of the view that the application could not be reinstated, as the grace period had elapsed.
The FCA affirmed Boivin J, who had upheld the Commissioner’s decision: "the relevant provisions of the Rules could not be clearer. Subsection 6(1) provides that “for the purpose of prosecuting or maintaining an application the Commissioner shall only communicate with, and shall only have regard to communications from, the authorized correspondent.” The wording of section 3.1, which deals with the late payment of fees, makes it clear that this prohibition extends to communications relating to all such payments as it operates ‘subject to subsection 6(1)’” [34]. The decision of Hughes J in Sarnoff Corp. v. Canada (Attorney General), 2008 FC 712, aff'd 2009 FCA 142 was persuasively distinguished on the basis that Hughes J, faced with ambiguous evidence, had found as a fact that the notice of appointment of agent had been received by CIPO [29, FC 20].
The Court’s analysis is entirely convincing as a matter of law, but the result is not satisfactory. As Hughes J pointed out in Sarnoff “The seemingly minor fault in having maintenance fees actually paid, received and recorded by a firm which, arguably, at the time was not the patent agent or associate agent of record, results in the wholly disproportionate result of loss of all rights to receive the grant of patent monopoly rights should the application otherwise prove to be acceptable.” [32]. It is true that the facts in Unicrop were different enough to compel a different result, but the point remains that the loss of all rights is disproportionate to the fault.
In Unicrop the FCA affirmed that maintenance fees must be paid by the authorized correspondent, and that the Commissioner is entitled, and indeed apparently required, to refuse payment from anyone else. It is difficult to argue with this conclusion, given the text of the Patent Rules; the question is why the Rules impose such a stringent requirement.
Unicrop’s representatives on filing were Bereskin & Parr LLP, who also paid the first two annual maintenance fees. The third annual fee was not paid in time and the application was deemed abandoned. Just prior to the one year grace period for reinstatement, the applicant attempted to reinstate the application through their new agent, Furman & Kallio. However, the notice of appointment of agent had not been submitted to CIPO prior to the grace period deadline [FC 27]. CIPO therefore refused to accept the fee payment, on the basis that it had not been submitted by the authorized correspondent. The Appointment of Agent form was submitted to CIPO after the deadline, but CIPO was of the view that the application could not be reinstated, as the grace period had elapsed.
The FCA affirmed Boivin J, who had upheld the Commissioner’s decision: "the relevant provisions of the Rules could not be clearer. Subsection 6(1) provides that “for the purpose of prosecuting or maintaining an application the Commissioner shall only communicate with, and shall only have regard to communications from, the authorized correspondent.” The wording of section 3.1, which deals with the late payment of fees, makes it clear that this prohibition extends to communications relating to all such payments as it operates ‘subject to subsection 6(1)’” [34]. The decision of Hughes J in Sarnoff Corp. v. Canada (Attorney General), 2008 FC 712, aff'd 2009 FCA 142 was persuasively distinguished on the basis that Hughes J, faced with ambiguous evidence, had found as a fact that the notice of appointment of agent had been received by CIPO [29, FC 20].
The Court’s analysis is entirely convincing as a matter of law, but the result is not satisfactory. As Hughes J pointed out in Sarnoff “The seemingly minor fault in having maintenance fees actually paid, received and recorded by a firm which, arguably, at the time was not the patent agent or associate agent of record, results in the wholly disproportionate result of loss of all rights to receive the grant of patent monopoly rights should the application otherwise prove to be acceptable.” [32]. It is true that the facts in Unicrop were different enough to compel a different result, but the point remains that the loss of all rights is disproportionate to the fault.
Thursday, February 17, 2011
Must the Factual Basis for Sound Prediction Be Disclosed in the Patent?
Eli Lilly Canada Inc. v. Apotex Inc. / raloxifene (NOC), 2008 FC 142, 63 CPR (4th) 406, aff'd 2009 FCA 97, 78 CPR(4th) 388.
In Eli Lilly / raloxifene (NOC), 2008 FC 142 Hughes J held that the data supporting a sound prediction of utility must be disclosed in the patent itself [163-64]. This was the primary basis for holding the allegation of invalidity to be justified [183]. The holding that the disclosure must be in the patent was crucial, as Hughes J held that a good factual basis for the sound prediction did exist, and if it had been disclosed in the patent, that ground of attack would have failed [156 - 58]. Hughes J’s holding on this point was specifically affirmed by the Court of Appeal in 2009 FCA 97 [15] Noël JA: Desjardins, Trudel JJA. In this post I will argue that despite this clear holding, the doctrine that the factual basis for sound prediction must be disclosed in the patent itself is conceptually unsound and inconsistent with the leading cases, including those of the Supreme Court of Canada. Recent jurisprudence from the FCA has provided a clarification that will allow the application of the doctrine to be avoided in many, perhaps most cases, but we nonetheless are left with an unjustifiable and technical doctrine that is uniquely Canadian, and so may serve as a trap for unwary patent drafters.
Taking the conceptual difficulty first, it is well established that the data supporting utility need not be disclosed in the patent itself. Indeed, the patentee need not even explain how the invention is useful: Consolboard v. MacMillam Bloedel, [1981] 1 SCR 504 at 526; Pfizer / atorvastatin calcium (NOC), 2008 FCA 108 [57]-[62]; Pfizer / sildenafil (NOC) 2010 FCA 242 [82]. Consequently, the raloxifene rule implies a sharp distinction between a patent where utility has been established at the relevant date, and one where utility is based on sound prediction. In the former case the patent need not disclose any data supporting utility, but in the latter it must: see 2010 FC 1065 [92]. The difficulty with this distinction is that it is not consistent with a purposive interpretation of the Patent Act. On the text of the Act, there is only one utility requirement, namely that which defines an “invention” to be “new and useful.” There is therefore no textual basis for treating demonstrated utility and sound prediction differently. The purpose of that utility requirement is to ensure that a patent is not granted for “mere speculation”: Wellcome / AZT 2002 SCC 77 [69]. A line must be therefore drawn somewhere in the development of an innovation from conception to commercialization. That line is drawn at the point of sound prediction. Thus there is no purposive basis for the distinction between demonstration of actual utility and sound prediction. An innovation that has demonstrated utility is closer to practical application than one where there is only a sound prediction of utility, and an innovation that has actually been delivered to consumers is closer again; but none are based on mere speculation, and therefore all are equally “useful” under the Act. Note that the Supreme Court in Wellcome / AZT consistently treated demonstrated utility and sound prediction on the same terms: [46], [52], [56].
In Eli Lilly / raloxifene (NOC), 2008 FC 142 Hughes J held that the data supporting a sound prediction of utility must be disclosed in the patent itself [163-64]. This was the primary basis for holding the allegation of invalidity to be justified [183]. The holding that the disclosure must be in the patent was crucial, as Hughes J held that a good factual basis for the sound prediction did exist, and if it had been disclosed in the patent, that ground of attack would have failed [156 - 58]. Hughes J’s holding on this point was specifically affirmed by the Court of Appeal in 2009 FCA 97 [15] Noël JA: Desjardins, Trudel JJA. In this post I will argue that despite this clear holding, the doctrine that the factual basis for sound prediction must be disclosed in the patent itself is conceptually unsound and inconsistent with the leading cases, including those of the Supreme Court of Canada. Recent jurisprudence from the FCA has provided a clarification that will allow the application of the doctrine to be avoided in many, perhaps most cases, but we nonetheless are left with an unjustifiable and technical doctrine that is uniquely Canadian, and so may serve as a trap for unwary patent drafters.
Taking the conceptual difficulty first, it is well established that the data supporting utility need not be disclosed in the patent itself. Indeed, the patentee need not even explain how the invention is useful: Consolboard v. MacMillam Bloedel, [1981] 1 SCR 504 at 526; Pfizer / atorvastatin calcium (NOC), 2008 FCA 108 [57]-[62]; Pfizer / sildenafil (NOC) 2010 FCA 242 [82]. Consequently, the raloxifene rule implies a sharp distinction between a patent where utility has been established at the relevant date, and one where utility is based on sound prediction. In the former case the patent need not disclose any data supporting utility, but in the latter it must: see 2010 FC 1065 [92]. The difficulty with this distinction is that it is not consistent with a purposive interpretation of the Patent Act. On the text of the Act, there is only one utility requirement, namely that which defines an “invention” to be “new and useful.” There is therefore no textual basis for treating demonstrated utility and sound prediction differently. The purpose of that utility requirement is to ensure that a patent is not granted for “mere speculation”: Wellcome / AZT 2002 SCC 77 [69]. A line must be therefore drawn somewhere in the development of an innovation from conception to commercialization. That line is drawn at the point of sound prediction. Thus there is no purposive basis for the distinction between demonstration of actual utility and sound prediction. An innovation that has demonstrated utility is closer to practical application than one where there is only a sound prediction of utility, and an innovation that has actually been delivered to consumers is closer again; but none are based on mere speculation, and therefore all are equally “useful” under the Act. Note that the Supreme Court in Wellcome / AZT consistently treated demonstrated utility and sound prediction on the same terms: [46], [52], [56].
Tuesday, February 15, 2011
Lovastatin Background
Merck & Co Inc. v. Apotex Inc. / lovastatin 2010 FC 1265 Snider J
I have now commented on a variety of the legal issues that were raised in Snider J’s 244 page lovastatin decision, and while those issues do stand alone, a factual background might be useful. Of course, it would have been better if I had provided this background at the outset, but I was anxious to post on the legal issues.
The sole patent in issue, #1,161,380 (granted in 1984 – old Act), was a process and product-by-process patent to lovastatin when made with the micro-organism A. terreus. Lovastatin, the first commercialized “statin” sold in the Canadian market for the treatment of elevated blood cholesterol, was sold by Merck as MEVACOR. Merck was unsuccessful in obtaining an order of prohibition under the PM(NOC) Regulations because the statutory stay expired before the application was heard on the merits. (In 2010 FC 1264, released together with 2010 FC 1265, Snider J held that Apotex was not entitled to section 8 compensation: see my posts here and here.) This means that Snider J faced the infringement and validity questions on a clean slate.
In anticipation of obtaining a compulsory licence, Apotex had developed an infringing process for the production of lovastatin, AFI-1, that used A. terreus. In anticipation of the compulsory licencing regime being replaced by the PM(NOC) Regulations, Apotex subsequently developed a non-infringing process, AFI-4, which used a different organism, C. fuckelii. Apotex supplied the Canadian market primarily with product made in China by Blue Treasure, a joint venture with Chinese partners. Apotex transferred know-how related to both the AFI-1 and AFI-4 technologies to Blue Treasure. Once it became clear that Apotex would proceed under the NOC Regulations rather than by compulsory licence, Apotex insisted that Blue Treasure use the non-infringing AFI-4 technology. However, this was more expensive than the infringing AFI-1 process, and Snider J ultimately held on the facts that Blue Treasure had been boosting its profits by using the cheaper infringing process to make the lovastatin that it delivered to Apotex.
An interesting practical point is that Snider J accepted evidence that traces of DNA of the producing organism were present in the final product, so that by identifying the DNA it was possible to determine which process had been used. Snider J relied on this evidence in holding that one particular batch, that had been made in Winnipeg, had been made by the infringing process [465]. This DNA evidence was crucial for the particular batch in question, as the documentary evidence was not conclusive. (Snider J did not rely on DNA evidence for the Blue Treasure product as the provenance of the samples was not proven adequately, and documentary evidence sufficiently established infringement.) While interesting, this does not appear to raise any novel questions of evidence law. The primary debate, as to whether traces of DNA could survive the processing and whether Merck’s testing was reliable, was purely factual.
I have already commented on most of the legal issues raised by the decision. For completeness, I will mention two more. Apotex had argued that Merck was not the first inventor of the compound as claimed: recall that at the time the Act embodied a first-to-invent system. However, the Act at the time, in section 61(1), placed certain limitations on the challenge to validity based on inventorship. After a rigorous interpretation of that section Snider J held that Apotex’s challenge based on inventorship was precluded by the section, and in any event, there was no conflict in inventorship. Finally, a certain quantity of product was exempted on the basis of the regulatory working exemptions in s 55.2(1) of the current Act. On the only contentious point, Snider J held that production of product for the purpose of improving Apotex’s manufacturing process was exempt, on a straightforward application of Micro Chemicals [1972] S.C.R. 506.
I have now commented on a variety of the legal issues that were raised in Snider J’s 244 page lovastatin decision, and while those issues do stand alone, a factual background might be useful. Of course, it would have been better if I had provided this background at the outset, but I was anxious to post on the legal issues.
The sole patent in issue, #1,161,380 (granted in 1984 – old Act), was a process and product-by-process patent to lovastatin when made with the micro-organism A. terreus. Lovastatin, the first commercialized “statin” sold in the Canadian market for the treatment of elevated blood cholesterol, was sold by Merck as MEVACOR. Merck was unsuccessful in obtaining an order of prohibition under the PM(NOC) Regulations because the statutory stay expired before the application was heard on the merits. (In 2010 FC 1264, released together with 2010 FC 1265, Snider J held that Apotex was not entitled to section 8 compensation: see my posts here and here.) This means that Snider J faced the infringement and validity questions on a clean slate.
In anticipation of obtaining a compulsory licence, Apotex had developed an infringing process for the production of lovastatin, AFI-1, that used A. terreus. In anticipation of the compulsory licencing regime being replaced by the PM(NOC) Regulations, Apotex subsequently developed a non-infringing process, AFI-4, which used a different organism, C. fuckelii. Apotex supplied the Canadian market primarily with product made in China by Blue Treasure, a joint venture with Chinese partners. Apotex transferred know-how related to both the AFI-1 and AFI-4 technologies to Blue Treasure. Once it became clear that Apotex would proceed under the NOC Regulations rather than by compulsory licence, Apotex insisted that Blue Treasure use the non-infringing AFI-4 technology. However, this was more expensive than the infringing AFI-1 process, and Snider J ultimately held on the facts that Blue Treasure had been boosting its profits by using the cheaper infringing process to make the lovastatin that it delivered to Apotex.
An interesting practical point is that Snider J accepted evidence that traces of DNA of the producing organism were present in the final product, so that by identifying the DNA it was possible to determine which process had been used. Snider J relied on this evidence in holding that one particular batch, that had been made in Winnipeg, had been made by the infringing process [465]. This DNA evidence was crucial for the particular batch in question, as the documentary evidence was not conclusive. (Snider J did not rely on DNA evidence for the Blue Treasure product as the provenance of the samples was not proven adequately, and documentary evidence sufficiently established infringement.) While interesting, this does not appear to raise any novel questions of evidence law. The primary debate, as to whether traces of DNA could survive the processing and whether Merck’s testing was reliable, was purely factual.
I have already commented on most of the legal issues raised by the decision. For completeness, I will mention two more. Apotex had argued that Merck was not the first inventor of the compound as claimed: recall that at the time the Act embodied a first-to-invent system. However, the Act at the time, in section 61(1), placed certain limitations on the challenge to validity based on inventorship. After a rigorous interpretation of that section Snider J held that Apotex’s challenge based on inventorship was precluded by the section, and in any event, there was no conflict in inventorship. Finally, a certain quantity of product was exempted on the basis of the regulatory working exemptions in s 55.2(1) of the current Act. On the only contentious point, Snider J held that production of product for the purpose of improving Apotex’s manufacturing process was exempt, on a straightforward application of Micro Chemicals [1972] S.C.R. 506.
Monday, February 14, 2011
Claim Construction and Inoperable Species
Merck & Co Inc. v. Apotex Inc. / lovastatin 2010 FC 1265, Snider J
As noted in my last post, claim 1 of the patent at issue in the lovastatin case related to a specified compound produced by the microorganism A. terreus. It was established [492] that not all stains of A terreus would produce the product in question. Apotex argued that the claim was therefore invalid for lack of utility [492]. Snider J rejected this argument on the basis that the claims, properly construed, included only those strains capable of producing lovastatin [494], [110]-[121].
This point of claim construction is important because of the rule that a claim lacks utility if “[t]ere is evidence of lack of utility in respect of some of the area covered” [495]. I argued in a previous post that this rule is unsound. However, it has been accepted and applied in two SCC decisions, namely Minerals Separation North American Corp. v. Noranda Mines Ltd. (1952) 69 R.P.C. 81 (J.C.P.C.) affm’g [1950] S.C.R. 36(SCC) and Société des Usines Chimiques Rhone-Poulenc v. Jules R. Gilbert Ltd [1968] S.C.R. 950. Consequently, rather than rejecting the rule itself, Canadian courts have largely avoided its application by construing the claims to exclude the inoperable species, as did Snider J. This approach was established by the SCC decision in Burton Parsons Chemicals, Inc. v. Hewlett-Packard (Canada) Ltd. [1976] 1 S.C.R. 555 which distinguished Minerals Separation and Rhone-Poulenc on the basis of claim construction.
In Burton Parsons, the claim in question, to an electrocardiograph cream, specified that a cream of specified composition that was “compatible with normal skin.” The SCC adverted to this, noting that in Minerals Separation and Rhone-Poulenc “the object of the patent was some substances of a definite chemical composition” and “[u]nfortunately for the patentees, the claims covered at the same time” some compounds which were not useful for the specified purposes. The Court also noted that the inutility of the compounds in the earlier cases “was not known to the prior art. This is totally unlike the undesirable properties of some highly ionizable salts which Hewlett-Packard listed as objectionable. Their noxious character was well known . . .”
Thus Burton Parsons left two points open. First, what are the limits of such functional phrases: is it necessary that the inoperable embodiments were known at the time? Second, is it necessary that the claim expressly state a functional limitation, such as “compatible with normal skill,” or can this be read in on the basis that a person skilled in the art would appreciate that requirement? There was certainly many statements in Burton Parsons that would imply a generous reliance on the understanding of a person skilled in the art, but the points on which the Court distinguished the prior cases suggest a narrower approach.
The first question was addressed in Apotex Inc v Lundbeck Canada Inc / escitalopram (NOC) 2010 FCA 320 (Noël JA: Pelletier, JJA) affm'g 2009 FC 146 (Harrington J), in which claim 1 claimed “non-toxic acid addition salts” of the specified compound. The pamoic acid salt was conceded to be toxic [105], and Apotex argued that claim 1 was therefore invalid, as there was no suggestion that pamoate salt was toxic as of the claim date. It appears to be very clear on the facts that the toxicity of the pamoate salt would not have been obvious on the claim, as the patentee has specifically claimed that salt in claim 2, which was held to be invalid for that reason. The FCA nonetheless held that claim to be valid as excluding non-toxic salts in its terms, including the pamoate acid salt [108]. It appears, therefore, that it is not essential that the functional limitation be known as of the claim date.
The second question was at issue in the lovastatin decision as “[n]either claim 1 nor the specification explicitly states that the '380 Patent excludes nonproducing strains of Aspergillus terreus” [111]. It was established on the facts that a skilled person would know that not every strain of A terreus would produce the desired results [117] and that it would be routine for a skilled person to identify and optimize the producing strains [120]. On these facts, Snider J had no difficulty finding that it was an implicit requirement that non-producing strains are excluded from the scope of the claim [121]. In support of this conclusion, she cited only the general principle from Consolboard [1981] 1 SCR 504, 520, that the patent should be read “being neither benevolent nor harsh but rather seeking a construction which is reasonable and fair to both patentee and public.” While Snider J’s decision was not unduly burdened with authorities on this point (it should be understood that this was a relatively minor part of a long and complex decision), general principles are enough to show that her holding is surely right: a patent must be approached with “a mind willing to understand, not by a mind desirous of misunderstanding” (Whirlpool v Camco 2000 SCC 67 [49(c)]). If more authorities are needed, we may note that her holding is consistent with the dictum in Burton Parsons that “In Sandoz Patents Ltd. v. Gilcross Ltd., we had no hesitation in upholding claims for "therapeutically tolerable salts" of thioridazine to be obtained by reacting "with a therapeutically acceptable acid". I cannot think that the omission of the qualification "therapeutically acceptable" would have voided the patent and I will note that in the Rhône-Poulenc case this question was left open.”
But if Snider J is right that the operability limitation can be implied, and the FCA was right in Lundbeck / escitalopram (NOC) that inoperability need not be known as of the claim date, then it becomes increasingly difficult to distinguish Minerals Separation and Rhone-Poulenc. I argued in my last post that the principle they stand for is unsound. It now seems evident that they would not be decided the same way today. Perhaps the time has come to recognize that they are no longer good law.
As noted in my last post, claim 1 of the patent at issue in the lovastatin case related to a specified compound produced by the microorganism A. terreus. It was established [492] that not all stains of A terreus would produce the product in question. Apotex argued that the claim was therefore invalid for lack of utility [492]. Snider J rejected this argument on the basis that the claims, properly construed, included only those strains capable of producing lovastatin [494], [110]-[121].
This point of claim construction is important because of the rule that a claim lacks utility if “[t]ere is evidence of lack of utility in respect of some of the area covered” [495]. I argued in a previous post that this rule is unsound. However, it has been accepted and applied in two SCC decisions, namely Minerals Separation North American Corp. v. Noranda Mines Ltd. (1952) 69 R.P.C. 81 (J.C.P.C.) affm’g [1950] S.C.R. 36(SCC) and Société des Usines Chimiques Rhone-Poulenc v. Jules R. Gilbert Ltd [1968] S.C.R. 950. Consequently, rather than rejecting the rule itself, Canadian courts have largely avoided its application by construing the claims to exclude the inoperable species, as did Snider J. This approach was established by the SCC decision in Burton Parsons Chemicals, Inc. v. Hewlett-Packard (Canada) Ltd. [1976] 1 S.C.R. 555 which distinguished Minerals Separation and Rhone-Poulenc on the basis of claim construction.
In Burton Parsons, the claim in question, to an electrocardiograph cream, specified that a cream of specified composition that was “compatible with normal skin.” The SCC adverted to this, noting that in Minerals Separation and Rhone-Poulenc “the object of the patent was some substances of a definite chemical composition” and “[u]nfortunately for the patentees, the claims covered at the same time” some compounds which were not useful for the specified purposes. The Court also noted that the inutility of the compounds in the earlier cases “was not known to the prior art. This is totally unlike the undesirable properties of some highly ionizable salts which Hewlett-Packard listed as objectionable. Their noxious character was well known . . .”
Thus Burton Parsons left two points open. First, what are the limits of such functional phrases: is it necessary that the inoperable embodiments were known at the time? Second, is it necessary that the claim expressly state a functional limitation, such as “compatible with normal skill,” or can this be read in on the basis that a person skilled in the art would appreciate that requirement? There was certainly many statements in Burton Parsons that would imply a generous reliance on the understanding of a person skilled in the art, but the points on which the Court distinguished the prior cases suggest a narrower approach.
The first question was addressed in Apotex Inc v Lundbeck Canada Inc / escitalopram (NOC) 2010 FCA 320 (Noël JA: Pelletier, JJA) affm'g 2009 FC 146 (Harrington J), in which claim 1 claimed “non-toxic acid addition salts” of the specified compound. The pamoic acid salt was conceded to be toxic [105], and Apotex argued that claim 1 was therefore invalid, as there was no suggestion that pamoate salt was toxic as of the claim date. It appears to be very clear on the facts that the toxicity of the pamoate salt would not have been obvious on the claim, as the patentee has specifically claimed that salt in claim 2, which was held to be invalid for that reason. The FCA nonetheless held that claim to be valid as excluding non-toxic salts in its terms, including the pamoate acid salt [108]. It appears, therefore, that it is not essential that the functional limitation be known as of the claim date.
The second question was at issue in the lovastatin decision as “[n]either claim 1 nor the specification explicitly states that the '380 Patent excludes nonproducing strains of Aspergillus terreus” [111]. It was established on the facts that a skilled person would know that not every strain of A terreus would produce the desired results [117] and that it would be routine for a skilled person to identify and optimize the producing strains [120]. On these facts, Snider J had no difficulty finding that it was an implicit requirement that non-producing strains are excluded from the scope of the claim [121]. In support of this conclusion, she cited only the general principle from Consolboard [1981] 1 SCR 504, 520, that the patent should be read “being neither benevolent nor harsh but rather seeking a construction which is reasonable and fair to both patentee and public.” While Snider J’s decision was not unduly burdened with authorities on this point (it should be understood that this was a relatively minor part of a long and complex decision), general principles are enough to show that her holding is surely right: a patent must be approached with “a mind willing to understand, not by a mind desirous of misunderstanding” (Whirlpool v Camco 2000 SCC 67 [49(c)]). If more authorities are needed, we may note that her holding is consistent with the dictum in Burton Parsons that “In Sandoz Patents Ltd. v. Gilcross Ltd., we had no hesitation in upholding claims for "therapeutically tolerable salts" of thioridazine to be obtained by reacting "with a therapeutically acceptable acid". I cannot think that the omission of the qualification "therapeutically acceptable" would have voided the patent and I will note that in the Rhône-Poulenc case this question was left open.”
But if Snider J is right that the operability limitation can be implied, and the FCA was right in Lundbeck / escitalopram (NOC) that inoperability need not be known as of the claim date, then it becomes increasingly difficult to distinguish Minerals Separation and Rhone-Poulenc. I argued in my last post that the principle they stand for is unsound. It now seems evident that they would not be decided the same way today. Perhaps the time has come to recognize that they are no longer good law.
Friday, February 11, 2011
Utility and Inoperable Species
Merck & Co Inc. v. Apotex Inc. / lovastatin 2010 FC 1265, Snider J
Claim 1 of the patent at issue in the lovastatin case (#1,161,380) related to a specified compound produced by the microorganism A. terreus. It was established [492] that not all stains of A terreus would produce the product in question. Apotex argued that the claim was therefore invalid for lack of utility [492]. Snider J rejected this argument on the basis that the claims, properly construed, included only those strains capable of producing lovastatin [494], [110]-[121].
This raises two questions. The most obvious is as to whether Snider J’s construction of the claim was sound, particularly because neither claim 1 nor the specification explicitly states that the '380 Patent excludes non-producing strains of A. terreus [111]. In other words, is it permissible to read in an operability limitation?
The more fundamental question concerns the underlying rule of law respecting utility. It is now routine in Canadian law to say that a claim lacks utility if “[t]here is evidence of lack of utility in respect of some of the area covered” by the claim [495], quoting Wellcome / AZT 2002 SCC 77 [56]. This rule can operate harshly. If an inventor discovers a new class of compounds that are generally useful, for example in treating a deadly disease, this rule implies that even if a single compound within that class does not have the promised utility, the claim is invalid, even though a person skilled in the art would have no difficulty identifying the inoperative species. This consequence may be avoided if the claims are construed to exclude inoperable species, as in Snider J’s lovastatin decision, but the claim construction issue cannot be properly understood without casting a critical eye on the underlying utility doctrine.
Claim 1 of the patent at issue in the lovastatin case (#1,161,380) related to a specified compound produced by the microorganism A. terreus. It was established [492] that not all stains of A terreus would produce the product in question. Apotex argued that the claim was therefore invalid for lack of utility [492]. Snider J rejected this argument on the basis that the claims, properly construed, included only those strains capable of producing lovastatin [494], [110]-[121].
This raises two questions. The most obvious is as to whether Snider J’s construction of the claim was sound, particularly because neither claim 1 nor the specification explicitly states that the '380 Patent excludes non-producing strains of A. terreus [111]. In other words, is it permissible to read in an operability limitation?
The more fundamental question concerns the underlying rule of law respecting utility. It is now routine in Canadian law to say that a claim lacks utility if “[t]here is evidence of lack of utility in respect of some of the area covered” by the claim [495], quoting Wellcome / AZT 2002 SCC 77 [56]. This rule can operate harshly. If an inventor discovers a new class of compounds that are generally useful, for example in treating a deadly disease, this rule implies that even if a single compound within that class does not have the promised utility, the claim is invalid, even though a person skilled in the art would have no difficulty identifying the inoperative species. This consequence may be avoided if the claims are construed to exclude inoperable species, as in Snider J’s lovastatin decision, but the claim construction issue cannot be properly understood without casting a critical eye on the underlying utility doctrine.
Wednesday, February 9, 2011
Protective Order and Executive Access to Confidential Information
Novopharm Limited v. Nycomed Canada Inc. / pantoprazole (NOC) 2011 FC 109 Mandamin J
When I started this blog I told myself that I would not comment on procedure cases, but I couldn’t help but notice that this case goes to the point I made in last week’s post that difficulties in discovery may arise in the context of lost profits damages, as well as in an accounting of profits.
In the action underlying this order, Novopharm is seeking damages against Nycomed under section 8 of the NOC Regulations. Novopharm brought a motion for a protective order restricting access to certain confidential financial material that Novopharm would have to produce in order to establish its damages. In particular, Novopharm was concerned that its financial information would be indirectly disclosed to Ranbaxy, with whom Nycomed had an authorized generic agreement for the drug in question. Novopharm requested that access be restricted to Nycomed’s designee who were not involved in the Ranbaxy agreement. Prothonotary Milczynski’s order granted access to several of Nycomed’s senior executives, including Nycomed’s CEO. Novopharm appealed.
While the decision was understandably brief, the points of principle relied on by Mandamin J suggest a strong presumption against restrictive access excluding the party’s executives, on the basis that it is the corporate executives who must make the key litigation decisions on the company’s behalf, and a client must be fully informed in order to properly instruct counsel: [36]-[37]. However, the more fully reasoned decisions appear to be more nuanced than Mandamin J’s brief discussion suggests. For example, while Warner-Lambert Co. v Glaxo Laboratories Limited, [1975] RPC 354 (CA) does state that “it is the corporate decision-makers who are authorized to make decisions on the company’s behalf,” the decision as a whole reflects a concern for the legitimate interests of both parties. According to Buckley LJ:
“the court is, in my opinion, confronted with a balance or conflict of expedients. . . .How can justice be done and at the same time effect be given to the rights of each party to the greatest possible extent? . . .In such a case a controlled measure of disclosure seems best calculated to serve the interests of justice. . . . His advisers are his agents in the matter, and strong grounds must be required for excluding the principal from knowledge which his agents properly acquire on his behalf. But this principle must be subject to some modification if trade secrets are to be protected from disclosure to possible competitors.”
In Warner-Lambert the Court of Appeal did grant access to the plaintiff’s CEO, but only after having explicitly considered whether the CEO could be properly advised without access to that information. The Court indicated that in a case, for example, where the evidence was of a technical nature such that the executive would take advice rather than form his own opinion, it would not be necessary to disclose to him the underlying facts. This principle was applied (without citing Warner-Lambert) in Automated Tabulation Inc. v. Canadian Market Images Ltd., 1995 CanLII 7073, 24 OR(3d) 292(ON SC) in which the plaintiff’s principal officers were precluded from access in part because “there were many available experts who were familiar with the relevant computer language who could be retained to assist the plaintiff” [4]. Conversely, where the technical nature of the information means that counsel cannot be properly instructed without input from the client, access is necessary: Murphy Oil Co. Ltd. v Predator Corp. Ltd., 2002 ABQB 992 [16]. Note that in Kimberly-Clark of Canada Ltd. v. Proctor & Gamble Co. et al (1989), 25 C.P.R.(3d) 12 (FCA), the Court of Appeal approved a very restrictive order that granted access to litigation counsel and “one designated representative of the receiving party . . . who is working directly on this litigation and to whom it is necessary that the material be disclosed for purposes of this litigation and who shall not be employed in research, development, or production of diaper products at the time of disclosures and for two (2) years after disclosure of such confidential information.” The Court varied this order by giving access to a patent agent who had been a member of the litigation team from the outset. However, the exclusion of executive officers was not challenged and the facts justifying the order as a whole were not given. A very helpful discussion of a range of authorities is found in the decision of the High Court of Ireland in Koger Inc v James O’Donnell [2009] IEHC 385. Kelly J concluded that a restriction that denying access to persons in the plaintiffs’ organisation “can be ordered but it is unusual. If such a restriction is to apply, there must be exceptional circumstances which would justify it.” On the facts, Kelly J held that “[t]he case is a finely balanced one,” and he consequently ordered limited disclosure of the material in question.
On the whole, it appears that a restrictive order of the type sought by Novopharm is exceptional, but the judicial analysis is not generally as one-sided as Mandamin J's general statements of principle might suggest.
With that said, on the facts Mandamin J’s decision is consistent with a more balanced approach. The factor he cited as most important [40] was that Nycomed is a small company in which all executives were involved with both strategic planning and the Ranbaxy agreement. The financial information on which damages are calculated are analogous to technical information at issue in an infringement action, and this implies that some executive from Nycomed would have to have access to the information if counsel were to be properly instructed. Novopharm acknowledged as much in that its proposed order would only have excluded designees involved in the Ranbaxy agreement. But on the facts, it was essentially impossible for Nycomed to satisfy that requirement while still providing proper instruction. A second significant point relied on by Mandamin J at [33]-[35] was that the documents Novopharm designated for restricted access were defined in very general terms, such as “commercial strategic planning materials for Novopharm’s or Nycomed’s pantoprazole products.” This is both broad and imprecise, and Mandamin J no doubt envisaged an endless series of motions seeking clarification.
Similarly, Murphy Oil Co. Ltd. v Predator Corp. Ltd., 2002 ABQB 992, relied on by Mandamin J, stated at [7] that "[t]he burden of justifying the need for an order precluding inspection of records by the parties themselves lies in the party seeking the order. Such an order is rarely given and the applicant’s burden is heavy." Despite this strong wording, it too is readily justified on a more balanced approach, as the plaintiff did not seek the order protecting its information until it had almost completed discovery of the equivalent information held by the defendant [14], and, as noted above, the information was technical and required input of the client to instruct counsel.
When I started this blog I told myself that I would not comment on procedure cases, but I couldn’t help but notice that this case goes to the point I made in last week’s post that difficulties in discovery may arise in the context of lost profits damages, as well as in an accounting of profits.
In the action underlying this order, Novopharm is seeking damages against Nycomed under section 8 of the NOC Regulations. Novopharm brought a motion for a protective order restricting access to certain confidential financial material that Novopharm would have to produce in order to establish its damages. In particular, Novopharm was concerned that its financial information would be indirectly disclosed to Ranbaxy, with whom Nycomed had an authorized generic agreement for the drug in question. Novopharm requested that access be restricted to Nycomed’s designee who were not involved in the Ranbaxy agreement. Prothonotary Milczynski’s order granted access to several of Nycomed’s senior executives, including Nycomed’s CEO. Novopharm appealed.
While the decision was understandably brief, the points of principle relied on by Mandamin J suggest a strong presumption against restrictive access excluding the party’s executives, on the basis that it is the corporate executives who must make the key litigation decisions on the company’s behalf, and a client must be fully informed in order to properly instruct counsel: [36]-[37]. However, the more fully reasoned decisions appear to be more nuanced than Mandamin J’s brief discussion suggests. For example, while Warner-Lambert Co. v Glaxo Laboratories Limited, [1975] RPC 354 (CA) does state that “it is the corporate decision-makers who are authorized to make decisions on the company’s behalf,” the decision as a whole reflects a concern for the legitimate interests of both parties. According to Buckley LJ:
“the court is, in my opinion, confronted with a balance or conflict of expedients. . . .How can justice be done and at the same time effect be given to the rights of each party to the greatest possible extent? . . .In such a case a controlled measure of disclosure seems best calculated to serve the interests of justice. . . . His advisers are his agents in the matter, and strong grounds must be required for excluding the principal from knowledge which his agents properly acquire on his behalf. But this principle must be subject to some modification if trade secrets are to be protected from disclosure to possible competitors.”
In Warner-Lambert the Court of Appeal did grant access to the plaintiff’s CEO, but only after having explicitly considered whether the CEO could be properly advised without access to that information. The Court indicated that in a case, for example, where the evidence was of a technical nature such that the executive would take advice rather than form his own opinion, it would not be necessary to disclose to him the underlying facts. This principle was applied (without citing Warner-Lambert) in Automated Tabulation Inc. v. Canadian Market Images Ltd., 1995 CanLII 7073, 24 OR(3d) 292(ON SC) in which the plaintiff’s principal officers were precluded from access in part because “there were many available experts who were familiar with the relevant computer language who could be retained to assist the plaintiff” [4]. Conversely, where the technical nature of the information means that counsel cannot be properly instructed without input from the client, access is necessary: Murphy Oil Co. Ltd. v Predator Corp. Ltd., 2002 ABQB 992 [16]. Note that in Kimberly-Clark of Canada Ltd. v. Proctor & Gamble Co. et al (1989), 25 C.P.R.(3d) 12 (FCA), the Court of Appeal approved a very restrictive order that granted access to litigation counsel and “one designated representative of the receiving party . . . who is working directly on this litigation and to whom it is necessary that the material be disclosed for purposes of this litigation and who shall not be employed in research, development, or production of diaper products at the time of disclosures and for two (2) years after disclosure of such confidential information.” The Court varied this order by giving access to a patent agent who had been a member of the litigation team from the outset. However, the exclusion of executive officers was not challenged and the facts justifying the order as a whole were not given. A very helpful discussion of a range of authorities is found in the decision of the High Court of Ireland in Koger Inc v James O’Donnell [2009] IEHC 385. Kelly J concluded that a restriction that denying access to persons in the plaintiffs’ organisation “can be ordered but it is unusual. If such a restriction is to apply, there must be exceptional circumstances which would justify it.” On the facts, Kelly J held that “[t]he case is a finely balanced one,” and he consequently ordered limited disclosure of the material in question.
On the whole, it appears that a restrictive order of the type sought by Novopharm is exceptional, but the judicial analysis is not generally as one-sided as Mandamin J's general statements of principle might suggest.
With that said, on the facts Mandamin J’s decision is consistent with a more balanced approach. The factor he cited as most important [40] was that Nycomed is a small company in which all executives were involved with both strategic planning and the Ranbaxy agreement. The financial information on which damages are calculated are analogous to technical information at issue in an infringement action, and this implies that some executive from Nycomed would have to have access to the information if counsel were to be properly instructed. Novopharm acknowledged as much in that its proposed order would only have excluded designees involved in the Ranbaxy agreement. But on the facts, it was essentially impossible for Nycomed to satisfy that requirement while still providing proper instruction. A second significant point relied on by Mandamin J at [33]-[35] was that the documents Novopharm designated for restricted access were defined in very general terms, such as “commercial strategic planning materials for Novopharm’s or Nycomed’s pantoprazole products.” This is both broad and imprecise, and Mandamin J no doubt envisaged an endless series of motions seeking clarification.
Similarly, Murphy Oil Co. Ltd. v Predator Corp. Ltd., 2002 ABQB 992, relied on by Mandamin J, stated at [7] that "[t]he burden of justifying the need for an order precluding inspection of records by the parties themselves lies in the party seeking the order. Such an order is rarely given and the applicant’s burden is heavy." Despite this strong wording, it too is readily justified on a more balanced approach, as the plaintiff did not seek the order protecting its information until it had almost completed discovery of the equivalent information held by the defendant [14], and, as noted above, the information was technical and required input of the client to instruct counsel.
Monday, February 7, 2011
What is a “patent”? Constitutionality of the Data Protection Regulations.
Apotex Inc. v. Canada (Health) 2010 FCA 334, Nadon JA: Sharlow, Layden-Stevenson JJA affm’g 2009 FC 725 Mandamin J
This case was a challenge to the validity of the Data Protection Regulations, on the basis, inter alia, of federalism. The heads of power advanced in support of validity were the criminal law power, trade and commerce, and POGG. Mandamin J held that the DPR is intra vires under the trade and commerce power [107], but not under the criminal law power [85]. The Court of Appeal held that it was valid under the criminal law power, and did not address whether it also fell under trade and commerce [132]. Neither addressed POGG [FC 110] [FCA 132].
I don't have the expertise to comment on the courts’ interpretation of the criminal law and trade and commerce powers. What puzzles me is that subsection 91(22) “Patents of Invention and Discovery,” was not even argued, though it seems to be the most natural head of power. Patents, after all, “are essentially about information as to what to make or do.”1 It is knowledge, not some physical thing, that provides the quid pro quo for which the patent is granted.2 A patentable invention is “the practical embodiment of the new knowledge.”3 All these descriptions of patents are equally applicable to the data at issue under the DPR, which is information as to the safety and efficacy of a particular drug. More generally, the purpose of patent law is to provide an incentive to disclose valuable new information that is costly to develop but easy to copy, by providing the patentee with a monopoly period at which its product can be priced above marginal cost, so allowing it to recoup development costs. Both the FCA at [109] and Mandamin J at [78], [83] agree that this is the purpose of the DPR. All this suggests to me that the DPR is a form of patent.
Of course, the DPR is not exactly like a modern patent granted under the Patent Act. Most obviously, there is no non-obviousness requirement. Protection under the DPR is most important precisely when the new drug is not innovative enough to be patentable under the Patent Act, and yet it is new enough to require new clinical testing prior to approval by the Department of Health. That the DPR does not require inventive ingenuity is irrelevant to whether it is a valid exercise of the patent power. Quite apart from any notion of the constitution as a “living tree,” at the time of confederation there was no requirement of non-obviousness recognized as such in patent law. The requirement for inventive ingenuity emerged gradually in the mid-to-late 19th century out of the novelty requirement. At the time of confederation the leading case was the decision of the House of Lords in Harwood v Great Northern Railway (1865) 11 E.R. 1488 holding at 1501 that “mere application of an old contrivance in the old way to an analogous subject [is] without novelty in the application.” Though this doctrine of analogous use eventually matured into the modern inventive step requirement, it was at that time still conceived of as an aspect of the novelty requirement. The exercise of inventive ingenuity as a separate requirement was not established as a distinct requirement until the 1880s.4 Even so, as late as the 5th edition in 1909, Terrell still referred to the issue as being one of novelty (at 37), and as late as 1947, in tracing the historical origins of the rquirement of inventive ingenuity, Dr. Fox argued that the whole doctrine was ill-advised and developed in error, and that the law should never have moved beyond the novelty requirements.5 Of course, Dr. Fox lost the day on this point, as he well knew, but his argument does show that inventive ingenuity cannot be considered to be a defining characteristic of patent law, either historically or normatively.
The core of the patent power, I would suggest, is to provide an incentive to the development of practical new knowledge that is difficult to develop and easy to copy, by providing a period of monopoly over the exploitation of that knowledge. Whether that definition is roughly correct, it is historically clear that non-obviousness is a relatively recent appendage, that must itself be justified by the living tree doctrine, not a central part of the concept. If anything, the protection provided by the DPR is closer to the conception of a patent that was current in 1867 than is our modern approach. Perhaps it might be said that the “living tree” metaphor works both ways, and though the DPR would have been within the patent power in 1867, it is no longer; the more obvious conclusion is simply that the DPR is valid under subsection 91(22).
1 Aerotel Ltd v Telco Ltd [2006] EWCA Civ 1371 [32] Jacob LJ.
2 Consolboard Inc v MacMillan Bloedel (Sask) Ltd, [1981] 1 SCR 504 at 517.
3 Shell Oil Co v Canada (Commissioner of Patents), [1982] 2 SCR 536 at 549.
4 See Hayward v Hamilton (1881) Griffin’s Patent Cases 115; Morgan v Windover (1887) 4 RPC 417 at 425 (Ch.D.); Gadd & Mason v Mayor etc. of Manchester (1892) 9 RPC 516 at 425-26 (CA).
5 Harold G. Fox, Monopolies and Patents: A Study of the History and Future of the Patent Monopoly (1947 Toronto, U of T Press), generally Ch. XVII and esp 237.
This case was a challenge to the validity of the Data Protection Regulations, on the basis, inter alia, of federalism. The heads of power advanced in support of validity were the criminal law power, trade and commerce, and POGG. Mandamin J held that the DPR is intra vires under the trade and commerce power [107], but not under the criminal law power [85]. The Court of Appeal held that it was valid under the criminal law power, and did not address whether it also fell under trade and commerce [132]. Neither addressed POGG [FC 110] [FCA 132].
I don't have the expertise to comment on the courts’ interpretation of the criminal law and trade and commerce powers. What puzzles me is that subsection 91(22) “Patents of Invention and Discovery,” was not even argued, though it seems to be the most natural head of power. Patents, after all, “are essentially about information as to what to make or do.”1 It is knowledge, not some physical thing, that provides the quid pro quo for which the patent is granted.2 A patentable invention is “the practical embodiment of the new knowledge.”3 All these descriptions of patents are equally applicable to the data at issue under the DPR, which is information as to the safety and efficacy of a particular drug. More generally, the purpose of patent law is to provide an incentive to disclose valuable new information that is costly to develop but easy to copy, by providing the patentee with a monopoly period at which its product can be priced above marginal cost, so allowing it to recoup development costs. Both the FCA at [109] and Mandamin J at [78], [83] agree that this is the purpose of the DPR. All this suggests to me that the DPR is a form of patent.
Of course, the DPR is not exactly like a modern patent granted under the Patent Act. Most obviously, there is no non-obviousness requirement. Protection under the DPR is most important precisely when the new drug is not innovative enough to be patentable under the Patent Act, and yet it is new enough to require new clinical testing prior to approval by the Department of Health. That the DPR does not require inventive ingenuity is irrelevant to whether it is a valid exercise of the patent power. Quite apart from any notion of the constitution as a “living tree,” at the time of confederation there was no requirement of non-obviousness recognized as such in patent law. The requirement for inventive ingenuity emerged gradually in the mid-to-late 19th century out of the novelty requirement. At the time of confederation the leading case was the decision of the House of Lords in Harwood v Great Northern Railway (1865) 11 E.R. 1488 holding at 1501 that “mere application of an old contrivance in the old way to an analogous subject [is] without novelty in the application.” Though this doctrine of analogous use eventually matured into the modern inventive step requirement, it was at that time still conceived of as an aspect of the novelty requirement. The exercise of inventive ingenuity as a separate requirement was not established as a distinct requirement until the 1880s.4 Even so, as late as the 5th edition in 1909, Terrell still referred to the issue as being one of novelty (at 37), and as late as 1947, in tracing the historical origins of the rquirement of inventive ingenuity, Dr. Fox argued that the whole doctrine was ill-advised and developed in error, and that the law should never have moved beyond the novelty requirements.5 Of course, Dr. Fox lost the day on this point, as he well knew, but his argument does show that inventive ingenuity cannot be considered to be a defining characteristic of patent law, either historically or normatively.
The core of the patent power, I would suggest, is to provide an incentive to the development of practical new knowledge that is difficult to develop and easy to copy, by providing a period of monopoly over the exploitation of that knowledge. Whether that definition is roughly correct, it is historically clear that non-obviousness is a relatively recent appendage, that must itself be justified by the living tree doctrine, not a central part of the concept. If anything, the protection provided by the DPR is closer to the conception of a patent that was current in 1867 than is our modern approach. Perhaps it might be said that the “living tree” metaphor works both ways, and though the DPR would have been within the patent power in 1867, it is no longer; the more obvious conclusion is simply that the DPR is valid under subsection 91(22).
1 Aerotel Ltd v Telco Ltd [2006] EWCA Civ 1371 [32] Jacob LJ.
2 Consolboard Inc v MacMillan Bloedel (Sask) Ltd, [1981] 1 SCR 504 at 517.
3 Shell Oil Co v Canada (Commissioner of Patents), [1982] 2 SCR 536 at 549.
4 See Hayward v Hamilton (1881) Griffin’s Patent Cases 115; Morgan v Windover (1887) 4 RPC 417 at 425 (Ch.D.); Gadd & Mason v Mayor etc. of Manchester (1892) 9 RPC 516 at 425-26 (CA).
5 Harold G. Fox, Monopolies and Patents: A Study of the History and Future of the Patent Monopoly (1947 Toronto, U of T Press), generally Ch. XVII and esp 237.
Thursday, February 3, 2011
Accounting Denied
Merck & Co Inc. v. Apotex Inc. / lovastatin 2010 FC 1265, Snider J
An accounting of profits is an equitable remedy and therefore discretionary, but it has been common practice in the Federal Court to allow a successful patentee to elect between an accounting and damages. In the lovastatin decision, Snider J denied an accounting to the patentee. The question is whether this is part of general reconsideration of the practice of allowing an election, as was called for by Hughes J in 2006 FC 524 [227], or whether it simply reflects the particular facts of the case.
In refusing to award the accounting Snider J adverted to two general points of principle. The first is that “[i]t is necessary for a party seeking an equitable remedy, such as profits, to show some basis for the exercise of equity,” [617], citing Janssen-Ortho Inc. v. Novopharm Ltd., 2006 FC 1234, 57 C.P.R. (4th) 58 at para. 132, (Hughes J). In Janssen-Ortho Hughes J denied an accounting for that reason alone. With respect, the view that equitable remedies are inherently extraordinary is unsound in principle and dangerous in practice. The historical reason for restricting access to equitable remedies was to restrict forum shopping and minimizing the case-load in equity, which had limited resources compared with the courts of law. In other words, it was not equitable remedies that were extraordinary, but access to the courts of equity. There is no good reason to treat equitable remedies as extraordinary after the unification of the courts: see my article “Interlocutory Injunctions and Irreparable Harm in the Federal Courts,” 88 Can Bar Rev 517 (2009). The issue here is not simply accounting of profits; permanent injunctions are also equitable remedies. While there may well be circumstances in which a permanent injunction is inappropriate (see eBay Inc. v. MercExchange (2006) 547 US 388), it surely would be inadequate to say that injunctions should routinely be denied simply because they are equitable in nature. The same is true for an accounting. The law has moved beyond such formalism.
The second point of principle is that “[a]n accounting of profits is not an easy calculation” [615]. While this is no doubt true, the relevant question is whether it is easier than damages. I may be missing something from my academic perch, but it is not clear to me why it is more difficult to calculate the defendant’s actual profits in an accounting than it is to calculate the plaintiff’s hypothetical lost profits in a damages assessment; certainly there are many well-known judicial statements remarking on the difficulty of calculating damages.* I can understand that a profits calculation of either kind might be more difficult than reasonable royalty damages, but a patentee is entitled to lost profit damages if it can show that it has lost sales, and the greater ease of calculating a reasonably royalty has never been a ground for confining the patentee to that remedy. One practical point that may make an accounting more difficult is that the defendant is liable to resist the necessary discovery to avoid disclosing business information to a competitor (Beloit v. Valmet OY (1992), 45 CPR(3d) 116 at 118 (FCA)). But it is disquieting to respond to this by instead requiring the successful patentee to disclose its business information to an infringing competitor. Of course, on the facts of a particular case it may be easier to calculate the plaintiff’s hypothetical lost profits than the defendant’s actual profits, but that is very different from saying that an accounting is generally more difficult.
The advantages of an accounting must also be recognized. One advantage stems from the very feature that gives rise to the difficulty: that it is the infringer, rather than the successful patentee that has to disclose its business information can be seen as an advantage of the remedy. But the more compelling point is that the alternative of a reasonable royalty, which is the easiest calculation, amounts to a compulsory licence. As Zinn J has pointed out in Monsanto Canada Inc. v. Rivett 2009 FC 317 at [23], “[a]t the level of principle, there is no deterrent from infringing the patent if what the infringer is required to hand over is the sum he would otherwise have paid to Monsanto to buy the seed and the licence. In fact, this would almost be counter to the purpose of deterrence. It is much like saying, as the plaintiffs put it in their oral submission, ‘Catch me if you can’. If caught, the defendant would be required to pay the sum he would have paid to use the patent in any event. When not caught, he is left with a windfall.”
Surely an accounting is a far better response to the deterrence problem than the US approach of treble damages for wilful infringement. The problem with the US approach is that determination of wilfulness is always uncertain; used too sparingly the deterrence is inadequate, but used too vigorously its punitive nature risks chilling legitimate activity. An accounting strikes a sound middle ground; the deterrence is adequate because the defendant is worse off than if it had obtained a licence; but the remedy is not punitive, in the sense that the defendant is no worse off for having infringed that if it had pursued some other activity entirely.
Given these advantages, I would suggest that the long-standing practice of permitting a successful patentee to elect an accounting, unless there is some reason to refuse on the facts, is sound. The difficulty of an accounting can be taken into account, but ideally it would be in comparison with the difficulty of calculating lost profits. The notion that an accounting somehow needs special justification simply because it is equitable in nature is an argument that should be banished.
On the facts, Snider J pointed to two factors that weighed against allowing an accounting. One was the delay in bringing the matter to trial and the consequent exceptional difficulty of calculation of the defendant’s profits [620]. Delay is a well accepted reason for refusing an equitable remedy, both generally and with respect to an accounting in particular, and to the extent that the patentee contributed to the delay, as Snider J suggested, then it is not unreasonable that the patentee should suffer this consequence. The second factor was the complexity of the defendant’s commercial arrangements, which would again increase the complexity of the calculation. Presumably, though Snider J did not say so directly, it would make calculation of Apotex’s profits more difficult than calculation of the plaintiffs’ lost profits. With that said, the implication, that a defendant may shield itself from an accounting by carefully structuring its supply arrangements, is somewhat troubling, though perhaps it is unrealistic to suppose that this consideration would be of sufficient practical importance to affect the defendant’s business structure.
In the end, I suggest that Snider J’s decision to deny an accounting is best seen as an application of the traditional approach to the particular facts of the case. Even though it has been common, even routine, to allow the patentee to elect an accounting, an accounting has never been considered as of right. There have always been occasional cases denying that remedy on the facts, and this is one of them.
* “[T]he whole subject-matter [of the calculation of damages] is one that is not capable of being mathematically ascertained by any exact figure.” Meters Ltd. v. Metropolitan Gas Meters Ltd. (1911), 28 RPC 157 at 166 (C.A.).
"[T]he restoration by way of compensation is therefore accomplished to a large extent by the exercise of a sound imagination and the practice of the broad axe." Watson Laidlaw Co. Ltd. v. Pott, Cassells and Williamson (1914), 31 RPC 104 at 117-118 (H.L.), cited and applied in Colonial Fastener Co. v. Lightning Fastener Co., [1937] SCR 36 at 44.
"[I]t is impossible to ascertain with arithmetical precision what in the ordinary course of business would have been the amount of the patentees' sales and profits." United Horse-Shoe & Nail Co. v. Stewart & Co. (1888), 5 RPC 260, L.R. 13 App.Cas. 401 at 413 (H.L.).
An accounting of profits is an equitable remedy and therefore discretionary, but it has been common practice in the Federal Court to allow a successful patentee to elect between an accounting and damages. In the lovastatin decision, Snider J denied an accounting to the patentee. The question is whether this is part of general reconsideration of the practice of allowing an election, as was called for by Hughes J in 2006 FC 524 [227], or whether it simply reflects the particular facts of the case.
In refusing to award the accounting Snider J adverted to two general points of principle. The first is that “[i]t is necessary for a party seeking an equitable remedy, such as profits, to show some basis for the exercise of equity,” [617], citing Janssen-Ortho Inc. v. Novopharm Ltd., 2006 FC 1234, 57 C.P.R. (4th) 58 at para. 132, (Hughes J). In Janssen-Ortho Hughes J denied an accounting for that reason alone. With respect, the view that equitable remedies are inherently extraordinary is unsound in principle and dangerous in practice. The historical reason for restricting access to equitable remedies was to restrict forum shopping and minimizing the case-load in equity, which had limited resources compared with the courts of law. In other words, it was not equitable remedies that were extraordinary, but access to the courts of equity. There is no good reason to treat equitable remedies as extraordinary after the unification of the courts: see my article “Interlocutory Injunctions and Irreparable Harm in the Federal Courts,” 88 Can Bar Rev 517 (2009). The issue here is not simply accounting of profits; permanent injunctions are also equitable remedies. While there may well be circumstances in which a permanent injunction is inappropriate (see eBay Inc. v. MercExchange (2006) 547 US 388), it surely would be inadequate to say that injunctions should routinely be denied simply because they are equitable in nature. The same is true for an accounting. The law has moved beyond such formalism.
The second point of principle is that “[a]n accounting of profits is not an easy calculation” [615]. While this is no doubt true, the relevant question is whether it is easier than damages. I may be missing something from my academic perch, but it is not clear to me why it is more difficult to calculate the defendant’s actual profits in an accounting than it is to calculate the plaintiff’s hypothetical lost profits in a damages assessment; certainly there are many well-known judicial statements remarking on the difficulty of calculating damages.* I can understand that a profits calculation of either kind might be more difficult than reasonable royalty damages, but a patentee is entitled to lost profit damages if it can show that it has lost sales, and the greater ease of calculating a reasonably royalty has never been a ground for confining the patentee to that remedy. One practical point that may make an accounting more difficult is that the defendant is liable to resist the necessary discovery to avoid disclosing business information to a competitor (Beloit v. Valmet OY (1992), 45 CPR(3d) 116 at 118 (FCA)). But it is disquieting to respond to this by instead requiring the successful patentee to disclose its business information to an infringing competitor. Of course, on the facts of a particular case it may be easier to calculate the plaintiff’s hypothetical lost profits than the defendant’s actual profits, but that is very different from saying that an accounting is generally more difficult.
The advantages of an accounting must also be recognized. One advantage stems from the very feature that gives rise to the difficulty: that it is the infringer, rather than the successful patentee that has to disclose its business information can be seen as an advantage of the remedy. But the more compelling point is that the alternative of a reasonable royalty, which is the easiest calculation, amounts to a compulsory licence. As Zinn J has pointed out in Monsanto Canada Inc. v. Rivett 2009 FC 317 at [23], “[a]t the level of principle, there is no deterrent from infringing the patent if what the infringer is required to hand over is the sum he would otherwise have paid to Monsanto to buy the seed and the licence. In fact, this would almost be counter to the purpose of deterrence. It is much like saying, as the plaintiffs put it in their oral submission, ‘Catch me if you can’. If caught, the defendant would be required to pay the sum he would have paid to use the patent in any event. When not caught, he is left with a windfall.”
Surely an accounting is a far better response to the deterrence problem than the US approach of treble damages for wilful infringement. The problem with the US approach is that determination of wilfulness is always uncertain; used too sparingly the deterrence is inadequate, but used too vigorously its punitive nature risks chilling legitimate activity. An accounting strikes a sound middle ground; the deterrence is adequate because the defendant is worse off than if it had obtained a licence; but the remedy is not punitive, in the sense that the defendant is no worse off for having infringed that if it had pursued some other activity entirely.
Given these advantages, I would suggest that the long-standing practice of permitting a successful patentee to elect an accounting, unless there is some reason to refuse on the facts, is sound. The difficulty of an accounting can be taken into account, but ideally it would be in comparison with the difficulty of calculating lost profits. The notion that an accounting somehow needs special justification simply because it is equitable in nature is an argument that should be banished.
On the facts, Snider J pointed to two factors that weighed against allowing an accounting. One was the delay in bringing the matter to trial and the consequent exceptional difficulty of calculation of the defendant’s profits [620]. Delay is a well accepted reason for refusing an equitable remedy, both generally and with respect to an accounting in particular, and to the extent that the patentee contributed to the delay, as Snider J suggested, then it is not unreasonable that the patentee should suffer this consequence. The second factor was the complexity of the defendant’s commercial arrangements, which would again increase the complexity of the calculation. Presumably, though Snider J did not say so directly, it would make calculation of Apotex’s profits more difficult than calculation of the plaintiffs’ lost profits. With that said, the implication, that a defendant may shield itself from an accounting by carefully structuring its supply arrangements, is somewhat troubling, though perhaps it is unrealistic to suppose that this consideration would be of sufficient practical importance to affect the defendant’s business structure.
In the end, I suggest that Snider J’s decision to deny an accounting is best seen as an application of the traditional approach to the particular facts of the case. Even though it has been common, even routine, to allow the patentee to elect an accounting, an accounting has never been considered as of right. There have always been occasional cases denying that remedy on the facts, and this is one of them.
* “[T]he whole subject-matter [of the calculation of damages] is one that is not capable of being mathematically ascertained by any exact figure.” Meters Ltd. v. Metropolitan Gas Meters Ltd. (1911), 28 RPC 157 at 166 (C.A.).
"[T]he restoration by way of compensation is therefore accomplished to a large extent by the exercise of a sound imagination and the practice of the broad axe." Watson Laidlaw Co. Ltd. v. Pott, Cassells and Williamson (1914), 31 RPC 104 at 117-118 (H.L.), cited and applied in Colonial Fastener Co. v. Lightning Fastener Co., [1937] SCR 36 at 44.
"[I]t is impossible to ascertain with arithmetical precision what in the ordinary course of business would have been the amount of the patentees' sales and profits." United Horse-Shoe & Nail Co. v. Stewart & Co. (1888), 5 RPC 260, L.R. 13 App.Cas. 401 at 413 (H.L.).
Subscribe to:
Posts (Atom)