Tuesday, February 7, 2006
Joel Schrag, an FTC economist, has posted an important paper on the effects of settlements in pharmaceutical patent suits on consumer welfare. Here are some important excerpts from the paper:
"A central question that I address is whether such settlements for “time off the patent” serve consumers’ interests. Given a choice, would consumers accept such settlements, or would they prefer that the parties litigate their cases? My main finding is that royalty-free licenses for delayed entry do not necessarily serve consumers’ interests, even if the entrants have all of the bargaining power in their negotiations with incumbent patent-holders and can therefore obtain relatively early entry dates through their settlements."
"I find that settlements for time off the patent generally leave consumers worse off than they would be if the firms litigated their cases. This conclusion does not change even if the terms of settlement are very favorable to the potential entrants, whose interests during settlement negotiations are to some degree aligned with consumers’ interests; both consumers and entrants prefer earlier entry."
"My analysis reveals two reasons that settlements can harm consumers’ interests. First, the time that consumers expect to benefit from a particular entrant represents a 1ower bound on the expected time that consumers would benefit from some entrant. For example, a particular entrant may have a fifty-fifty chance of winning an infringement case, but a settlement that enables the entrant to sell its product halfway through the remaining patent life could shortchange consumers, because another entrant may emerge if the first loses its infringement case. Given this possibility, consumers would have a greater than fifty-fifty chance of benefiting from some entrant. "
"Second, the form of patent settlement that I analyze can undermine the entrants’ incentives to invent around the incumbent’s patent. If the entrant that completes its product development first always settles with the incumbent, a slower entrant would likely never have a chance to earn duopoly profits. If the entrant that completes its product development first always litigates, meanwhile, a slower entrant would possibly have an opportunity to earn duopoly profits, because the first entrant may lose its patent infringement case. The prospect of settlement thus increases the relative importance to an entrant of being the first to develop a competing product, and, therefore, discourages an entrant from undertaking a project that it believes will take a long time to complete."
" The formal analysis does, however, support two broad conclusions. First, the model highlights the importance of analyzing the effects of any patent settlement on the incentives of the relevant third parties before drawing any conclusions about the settlement’s effect on consumers’ interests."
" Second, because the model reveals that patent settlements that license entry at a date that is consistent with the expected outcome of the trial are not necessarily beneficial to consumers, it suggests that other settlement terms that help parties reach agreement on a delayed entry license cannot be assumed to be ancillary to a pro-competitive outcome. If these other terms raise independent competition concerns, a settlement is even less likely to benefit consumers and should therefore be subject to even greater scrutiny."
On February 1, 2006, a Federal Circuit panel reconsidered and then vacated its earlier decision in Purdue Pharma v. Endo Pharmaceuticals, which had affirmed the trial court’s finding of inequitable conduct against Purdue. Purdue owns patents protecting its controlled release oxycodone product, Oxycontin®, a powerful pain reliever. Endo sought to manufacture a generic bioequivalent of Oxycontin®. Upon rehearing, the Federal Circuit panel vacated the trial court’s original opinion on inequitable conduct and remanded consideration of that issue for further proceedings consistent with the new Federal Circuit opinion. The Federal Circuit also affirmed the trial court’s finding of infringement, not particularly surprising when the allegedly infringing product is a bioequivalent of the patented product.
The key issue on appeal is whether the patentee, Purdue, committed inequitable conduct before the PTO. To establish inequitable conduct, a patent defendant must prove by clear and convincing evidence that the patentee intentionally made material misrepresentations to the PTO. The court considers materiality and intent on a sliding scale.
To overcome prior art before the PTO, Purdue had repeatedly implied and suggested, but did not directly state, that Purdue had clinical evidence proving that its patented product “controlled pain over a four-fold range of dosages for 90% of patients. . . .” Purdue had no such clinical evidence. Upon reconsideration, the panel found the materiality of Purdue’s indirect misrepresentations to be “not especially high . . . not as material as an affirmative misrepresentation would have been.” When the level of materiality is low, however, then intent cannot simply be inferred in a sliding scale analysis.
Direct evidence of intent is rarely available. The trial court relied on internal documents in which Purdue personnel admitted that they could not prove to the FDA with clinical evidence that its “oxycondone formulations were effective over a four-fold dosage range, compared to an eight-fold dosage range for other opioids.” But Purdue’s internal admissions that it lacked the quantum of proof necessary to obtain FDA approval are not sufficient to prove intent to deceive the PTO. “[E]vidence that Purdue personnel believed it would be difficult to satisfy FDA requirements is at best marginally related to whether they intended to deceive the PTO.”
Rather than reversing the trial court, the panel remanded the case, ordering the trial court to “rethink the relevance of the evidence relating to whether Purdue could prove that Oxycontin was the most easily titratable [capable of gradually adjusting the dosage to obtain the desired effect] analgesic.” Additionally, the trial court must reweigh its sliding scale analysis, “keep[ing] in mind that when the level of materiality is relatively low, the showing of intent must be proportionately higher.” As a result, to find inequitable conduct on remand, the trial court must find direct evidence of intent, distinct from the evidence on which it relied in its original opinion.