First published in LES Insights

Determining a reasonable royalty after a finding of patent infringement often involves complex considerations relating to the nature of the relevant markets. One important consideration may be the ability of the infringer to sell a noninfringing alternative, particularly if the market in question is highly regulated and has significant barriers to entry. In AstraZeneca AB v. Apotex Corp.,the District Court for the Southern District of New York held that the reasonable royalty for an infringing generic drug formulation was 50% of the infringer's profits. The court examined both the infringer and the patentee's positions at the time infringement began and determined that this royalty was reasonable because the infringer lacked a noninfringing formulation, the patentee did not license its patents, and the infringer's generic drug would have adversely affected the patentee's branded drug's market share and revenue.

Background

AstraZeneca developed and patented a compound called omeprazole to treat gastrointestinal conditions. Omeprazole is one of a class of drugs called proton-pump inhibitors, or PPIs. AstraZeneca patented both the omeprazole compound and formulations containing it. AstraZeneca's formulation patents solved several known issues with omeprazole, such as delivering the drug safely to the small intestine and protecting it from the acidic environment of the stomach. AstraZeneca's formulation included a core containing the active pharmaceutical, a stabilizing Alkaline Reacting Compound—an outer enteric coating to protect the core—and a water-soluble inter subcoating to separate the core from the enteric coating. It sold the patented formulation as the prescription drug Prilosec. The patent to omeprazole compounds expired in 2001, but the patent to formulations continued until 2007. Apotex manufactures and sells generic pharmaceuticals. In November 2003, it began selling a generic omeprazole formulation. AstraZeneca sued Apotex and the district court found that Apotex infringed the patent. This decision relates to the damages arising from that infringement.

The Decision

Before determining the reasonable royalty, the district court analyzed the market for PPI drugs and, in particular, the omeprazole market as of November 2003, when Apotex's infringement began. The court observed that in November 2003, several generic companies had entered the generic omeprazole market, competing with AstraZeneca's Prilosec drug. The court addressed AstraZeneca's claims in two waves and, in the first, determined that a formulation from Andrx infringed while a formulation from KUDCo did not. The second wave included formulations from Mylan, Lek, and Apotex. Thus, uncertainty surrounded the launch of those formulations because they were "at risk"of infringement. Notwithstanding that uncertainty, the infringing formulation from Andrx had a subcoating that formed in situ as the drug passed through the body, similar to Apotex's formulation.

The district court analyzed the effect of the generic formulations' launch on the PPI market and determined that they negatively affected AstraZeneca's sales of Prilosec. The court observed that a license from AstraZeneca to Apotex in November 2003 would have caused further price and market-share erosion of the PPI market because Apotex would have been able to price its drug aggressively to compete with AstraZeneca's Prilosec and the generic omeprazole drugs on the market.

The district court noted, however, that AstraZeneca anticipated this loss of market share after the expiration of its omeprazole patent and sought to position its next PPI drug, Nexium, as a superior drug to Prilosec and other PPIs on the market. As Prilosec's market share declined, Nexium's share grew from its introduction in 2001 through November 2003, when Apotex began infringing. AstraZeneca developed a comprehensive strategy in which Proctor & Gamble, under an agreement with AstraZeneca, launched an over-the-counter PPI, Prilosec OTC. If patients could not achieve satisfactory results from Prilosec OTC, AstraZeneca positioned Nexium as the alternative treatment to omeprazole-based Prilosec OTC. Despite AstraZeneca's positioning of Nexium relative to Prilosec OTC and other generics, price competition came as third-party payers (TPPs) such as managed-care companies tried to reduce their costs. AstraZeneca responded to this price competition by offering rebates on Nexium to promote its use and acceptance among TPPs. As more generic omeprazole drugs entered the market, AstraZeneca offered higher rebates to continue Nexium's growth in the PPI market.

The district court also looked to AstraZeneca's history of licensing its omeprazole patents, concluding the AstraZeneca generally did not license its patents in November 2003. The court examined several licenses to Takeda Chemical Industries, Byk Gulden Lomberg Chemische Fabrik, and Eisai Company, which resulted from litigation settlements, and concluded that none of these permitted the licensees to sell omeprazole. The court also examined AstraZeneca's agreement with Proctor & Gamble to sell Prilosec OTC, noting that AstraZeneca viewed the over-the-counter market as fundamentally different from the prescription-drug market that Apotex entered. The court also observed that Prilosec OTC served to further AstraZeneca's marketing strategy for Nexium and that Prilosec OTC used a slightly different active ingredient, omeprazole magnesium, rather than omeprazole.

The court also analyzed Apotex's ability to develop a noninfringing alternative in November 2003, noting that Apotex did not have a noninfringing product developed and would need at least two years to develop one. The court observed that one key factor in a hypothetical negotiation would have been Apotex's potential revenue loss from walking away from a license to develop its own noninfringing product. This delay was estimated to result in a 59% decrease in Apotex's profits. The court rejected Apotex's assertions that it could have avoided infringement by changing its composition in any one of three different ways, including developing a pellet formulation, copying other noninfringing products, or developing a microtablet formulation. The court observed that Apotex had attempted to develop noninfringing products since 1996, but had not developed any noninfringing alternatives by November 2003. Apotex had, instead, cancelled its microtablet project due to disappointing progress. The court, Apotex, and AstraZeneca all noted that, even if Apotex developed a noninfringing product quickly, it would still have to get FDA approval for the new formulation, which would result in delays and loss of revenue. In view of Apotex's failure to develop a noninfringing product between 1996 and 2003, the district court concluded that Apotex would have "had no reasonable confidence in the Fall of 2003 that it could find a noninfringing formulation for omeprazole that would obtain FDA approval, much less one that could be developed, tested, and approved without years of effort."

The court also rejected Apotex's position that it could have copied other noninfringing generic omeprazole compounds on the market. In November 2003, at the time of the hypothetical negotiation, only KUDCo manufactured a noninfringing generic omeprazole compound. But KUDCo had patented its formulation, and Apotex did not demonstrate that it could have copied this formulation without infringing KUDCo's patent. Two other generic companies, Lek and Mylan, were later found to provide noninfringing formulations, but Apotex would not have known that in November 2003. Moreover, the court noted that both Mylan and Lek had patented their formulation and Apotex would have infringed those patents by copying either of their products.

The court next turned to examining the information following Apotex's entry into the PPI market and considered both Nexium's and Prilosec OTC's effects on the market. First, Nexium had developed a significant market share in the prescription PPI market. Prilosec OTC, although significantly affecting the prescription omeprazole market, appeared to be helping AstraZeneca promote Nexium. Apotex's infringing product also averaged a 75% profit margin between November 2003 and October 2007, making it Apotex's highest gross-margin product.

Finally, the court looked at AstraZeneca's settlement offers with other generic manufacturers. It first looked at a settlement offer from Andrx, which offered between a 50% and 70% royalty to license AstraZeneca's patents. AstraZeneca rejected this offer. AstraZeneca also entered into a settlement with Teva, in which Teva paid a flat sum of less than $10 million. Although Apotex and AstraZeneca disputed how to interpret this settlement, the court noted that Teva paid 54% of its profits from its infringing sales.

Against this background, the district court tried to determine the reasonable royalty for Apotex's infringement. The court rejected Apotex's argument that the royalty should be based only on the value of the subcoating, not on the entire omeprazole formulation. The court noted that AstraZeneca's patent "substantially create[d] the value"of Apotex's infringing product because it solved several problems with developing an effective drug using omeprazole, and therefore, a royalty on the entire drug was appropriate.

The court then rejected Apotex's assertion that a royalty calculation should cease at the patent's expiration in April 2007, rather than at the end of the extended "pediatric exclusivity"period, which ended in October 2007. The district court rejected Apotex's argument that the Supreme Court's decision in Brulotte v. Thys2 precluded AstraZeneca from receiving a royalty beyond the patent's expiration date. The court concluded that Brulotte predated that pediatric-exclusivity statute and was therefore inapplicable, and also that Brulotte's concerns that a patentee would use a license to extend a monopoly was inapplicable where Congress specifically extended a patent's exclusionary period by statute.

As a result of this evidence, the court determined that the reasonable royalty would be 50% of Apotex's profits from its infringing sales from November 2003 (when Apotex's infringement began) through October 2007 (the end of the "pediatric-exclusivity"period). Although Apotex argued that would not have agreed to a 50% royalty, the court noted that Apotex's profit margins typically ranged from 31% to 48%, and that it estimated an omeprazole profit margin of 92%. Thus, even accounting for a royalty of 50% of Apotex's profits would still provide a net 46% profit to Apotex, which would fall at the upper end of its profit range. Furthermore, a license from AstraZeneca would have negated any concerns Apotex would be entering the market with an "at risk"product, allowing it to competitively price its product below its competitors to maximize market share and profit.

The court also determined that Apotex's inability to develop or use noninfringing formulations weighed in favor of a high royalty payment. The court noted that an optimistic two-year development cycle for a noninfringing product would deprive Apotex of an estimated 59% of its profit through delayed entry into the market. Apotex's inability to develop a noninfringing product and the likely delays and difficulties in developing one would have weighed in favor of obtaining a license and accepting the higher royalty payment.

The court next looked to AstraZeneca's position at the time of the hypothetical negotiation, determining that there was "overwhelming evidence"that AstraZeneca "was in the driver's seat in the negotiation and would have required Apotex to pay a hefty portion of its profits for a license."First, AstraZeneca had no program to license its patents and had shown no interest in licensing its patents for generic formulations. Second, if AstraZeneca did license its patents, it would have ensured that the license would have compensated it for any economic harm incurred from the entry of another generic omeprazole formulation into the market. The court noted that AstraZeneca feared that a licensee would significantly erode the price of prescription omeprazole and also negatively affect the PPI market as a whole, including negatively affecting AstraZeneca's Nexium strategy. In light of AstraZeneca's considerations, the court concluded that it would not have accepted a royalty of less than of 50% of Apotex's profits.

Although the court acknowledged that there "are no perfect benchmarks"for this royalty, it observed that a 50% profit royalty "fits comfortably"within the range of negotiations in the record. For example, AstraZeneca received a 40% royalty from Proctor & Gamble for Prilosec OTC, Andrx had offered AstraZeneca between 50% and 70% of its profits for a license, and Teva had paid a settlement amounting to 54% of its profit. Furthermore, the licenses of record from Proctor & Gamble, Andrx, and Teva, showed manufacturers' willingness to pay 50% or more of their profits to license AstraZeneca's patents.

Strategy and Conclusion

The AstraZeneca decision provides another example of applying several well-established factors for determining a reasonable royalty with a particular emphasis on the defendant's lack of a noninfringing alternative. In markets with high barriers to entry or significant regulatory hurdles that may result in monetary investment and time delays, this decision shows that high royalty rates may be set because the infringer would have, hypothetically, accepted a higher royalty to avoid delayed market entry, loss of market share, and loss of overall profit.