Canada’s Intellectual Property Firm

Federal Court of Appeal rules on non-infringing alternatives and apportionment as defences to an accounting of profits from patent infringement

On February 2, 2017, the Federal Court of Appeal released a significant decision on accounting for profits, a remedy for patent infringement in Canada: Apotex Inc v ADIR, 2017 FCA 23. The appeal concerned two defences raised by Apotex (and a related company) to reduce the profits it had to disgorge for its wrongful export sales of generic perindopril (COVERSYL).  The two defences - non-infringing alternatives (“NIAs”) and apportionment - are closely related, both contending that an infringer’s profits, in full or part, are not causally attributable to the infringement.

The Court of Appeal found that, in this circumstance, the Federal Court had erred in law by rejecting Apotex’s argument that the availability of NIAs reduced the profits to be disgorged.  The question of whether an NIA was available on the facts was remitted to the Federal Court. The Court of Appeal upheld the Federal Court’s decision not to apportion Apotex’s foreign revenues between those received for perindopril and those allegedly received for non-infringing services.


Starting around 2006, Apotex sold, in Canada and abroad, generic perindopril that it had manufactured in Canada. In 2008, as previously reported, the Federal Court found that these activities infringed ADIR’s Canadian Patent No. 1,341,196 (“196 patent”): 2008 FC 825, aff’d 2009 FCA 222.  The plaintiffs elected to claim an accounting of Apotex’s profits.

In 2015, after a 17-day hearing, a trial judge of the Federal Court determined the amount of Apotex’s profits attributable to infringing activity:  2015 FC 721 (previously reported here). At trial, Apotex acknowledged that to make domestic sales of perindopril it had to infringe the 196 patent; it followed that profits from these sales had to be completely disgorged.  At issue at trial, and on appeal, were Apotex’s profits from its export sales, particularly to Apotex affiliates in Australia and the UK.  Apotex received revenue from these affiliates under transfer price agreements.

Non-infringing alternatives

At trial, Apotex tried to establish that it could have made profits by selling NIAs to its foreign affiliates, namely perindopril active pharmaceutical ingredient and finished tablets manufactured outside Canada. In some of the hypothetical scenarios advanced by Apotex, selling NIAs would have been even more profitable than selling its real-world perindopril, ostensibly reducing to zero the profits from exports Apotex would have to disgorge.  The trial judge rejected the relevance at law of the alleged NIAs.

The Court of Appeal reversed.  The Court acknowledged that the Supreme Court of Canada in Monsanto Canada v Schmeiser, 2004 SCC 34, had found that the differential profit approach (“profits are allocated according to the value contributed to the defendant’s wares by the patent .... A comparison is to be made between the defendant’s profit attributable to the invention and his profit had he used the best non-infringing option”) was the preferred means of calculating an accounting of profits, and not the only means.   But the Court noted that a patentee should only receive that portion of the infringer’s profit that is causally attributable to the invention.  In this circumstance, the invention’s value could only be quantified if NIAs were considered, because a patent’s value lies in the ability to exclude competition.

The Court of Appeal provided guidance on how to determine whether an alleged NIA was available.  The Court observed that in considering this question, one enters a hypothetical world where the infringer did not infringe.  The Court held that its recent findings on the nature of the hypothetical world applied to an accounting of profits context.  In particular, an infringer must establish that in the hypothetical world it “could have” and “would have” obtained and used the NIA.  To prove “could have”, the infringer must demonstrate that it was possible for it to secure the NIA; to prove “would have”, the infringer must demonstrate that events would have transpired in such a way as to put them in that position.  See Apotex Inc  v Merck & Co, Inc, 2015 FCA 171 (a claim for damages for patent infringement, previously reported here) and Pfizer Canada Inc v Teva Canada Ltd, 2016 FCA 161 (a claim for damages under section 8 of the Patented Medicines (Notice of Compliance) Regulations, previously reported here).

The Court added that the inquiry is not limited to whether a supply of the NIA was available to replace the initial infringing sales, as the supply may become available at some later time in the hypothetical world.

The Court of Appeal remitted to the Federal Court the factual issue of whether Apotex could have and would have obtained non-infringing perindopril from three specific offshore suppliers, and, if so, whether Apotex could have and would have sold this material to its Australian and UK affiliates.


At trial, Apotex asserted that a portion of the transfer price paid by its Australian and UK affiliates for perindopril was for services that did not infringe the 196 patent, namely an indemnity for liability under foreign patents and related legal services. Apotex sought a $22 million deduction from its foreign profits on this basis.  The trial judge found that although apportionment could have been appropriate, Apotex had not led sufficient evidence.

The Court of Appeal upheld the decision not to apportion. The Court found that apportionment was not appropriate because “but for” the infringing perindopril, Apotex would not have earned anything on its sales to its Australian and UK affiliates.  What drove these sales were the new and useful characteristics of perindopril.  Thus, the profit resulting from these sales was entirely causally attributable to the patented invention. This finding was fatal to Apotex’s apportionment argument.  Nonetheless, the Court went on to also find that Apotex’s evidence for apportionment was insufficient, particularly as the transfer price agreements and surrounding factual matrix did not support an intention to provide the non-infringing services in exchange for a higher price for perindopril.

Finally, the Court of Appeal noted that Apotex’s apportionment argument had no air of reality.  It was not credible that Apotex objectively intended to apportion the transfer price where “this notion was first floated on the eve of trial”.

For further information, please contact a member of our firm’s Pharmaceutical group.

The preceding is intended as a timely update on Canadian intellectual property and technology law. The content is informational only and does not constitute legal or professional advice. To obtain such advice, please communicate with our offices directly.