Purdue Pharma v Epic Pharma

By: Gideon Eckhouse & Andreas Baltatzis  |   February 29, 2016

Purdue Pharma L.P. et al v. Epic Pharma LLC et al.

Appeal Nos. 2014-1294, -1296, -1306, -1307, -1312, -1313, -1314 (Fed. Cir. February 1, 2016)

In Purdue Pharma L.P. v. Epic Pharma, LLC, the Court of Appeals for the Federal Circuit reviewed the validity of multiple patents with claims directed toward pain-relieving dosage forms. In particular, the claims at issue were directed toward formulations containing the opioid analgesic oxycodone with a low amount of a specified impurity. The Court held that the patents are invalid as obvious or anticipated, despite the fact that there was no proof that the source of the impurity was known in the prior art. The Court also found that the hardness of the abuse-deterrent dosage form was inherent in the prior art.

Purdue asserted three patents with claims directed toward formulations containing oxycodone with a low amount of 14-hydroxycodeinone (“14-OH”). 14-OH is an impurity that is produced when oxycodone is made from poppy straw. The FDA asserted that 14-OH is a potentially carcinogenic impurity and mandated that Purdue, or any other supplier of an oxycodone formulation, lower the amount of 14-OH to less than 10 ppm. Purdue, through its subsidiary, Rhodes Technologies, developed an alternative method for producing oxycodone that resulted in lower levels of 14-OH. Purdue and Rhodes found that 14-OH was derived from an intermediate compound known as 8α.

In a separate application, Purdue attempted to patent a process for producing oxycodone with low levels of 14-OH. The process claims did not specify that the 8α intermediate was responsible for the production of the 14-OH that was removed in the process. In a separate, non-precedential decision, the Court found those claims to be invalid as obvious. See Chapman v. Casner, 315 F. App’x 294, 295 (Fed. Cir. 2009) (Rader, CJ., Dissenting). Purdue then applied for the patents-in-suit, which were directed toward the product, i.e., oxycodone with low levels of 14-OH, wherein the 14-OH is derived from 8α.

The District Court for the Southern District of New York found that the claims of the patents were invalid as obvious because a person of skill in the art would have known that: (1) 14-OH was produced during the preparation of oxycodone from poppy straw and (2) 14-OH could be removed from oxycodone through hydrogenation. The district court further held that the source of the 14-OH, i.e., 8α, was not relevant in the obviousness analysis because hydrogenation can be used to remove 14-OH, regardless of its source.

On appeal, Purdue argued that the district court erred in disregarding the 8α limitations from the claims. Purdue asserted that its scientists discovered the role of 8α in the production of the 14-OH impurity. Purdue relied on the precedent of Eibel Process Co. v. Minnesota & Ontario Paper Co., 261 U.S. 45 (1923), for the proposition that the discovery of the source of a problem is relevant in an obviousness determination for claims directed toward a solution to that problem. The panel of Chief Judge Prost, Judge Reyna, and Chief District Judge Leonard P. Stark of Delaware explained that Purdue’s reliance on the precedent was “misplaced.” According to the Court, even if the discovery of the role of 8α was not obvious, the claims were obvious because the amount of 14-OH can be reduced through hydrogenation, regardless of its source. Since the claims were only directed toward the end product, i.e., oxycodone with low levels of 14-OH, and not the means of solving a nonobvious problem, the claims would have been obvious.

The Court also upheld the lower court’s findings that the limitation that the 14-OH was “derived from 8α” was immaterial. The Court found that the “derived from 8α” limitation was a process limitation in a composition of matter claim. The specification disclosed the reduction of 14-OH regardless of its source. Purdue’s experts also testified that the structure of 14-OH does not change based on the source. Therefore, the Court held that the “derived from 8α” limitation cannot be a structural limitation. Since the claims included a product-by-process limitation, the Court found that the district court did not err in disregarding the limitations. The Court also reviewed secondary considerations, including alleged commercial success. Since Rhodes Technologies’ only “significant” consumer was Purdue, any evidence of commercial success was insufficient. The district court found, and the Federal Circuit agreed, that Purdue purchased oxycodone from Rhodes because it was a subsidiary, and therefore offered the product at a low price. Purdue’s decision to purchase oxycodone from Rhodes was not driven by the low levels of 14-OH or any other feature of the claims.

Purdue asserted a fourth patent, with claims directed toward an oxycodone formulation of a particular hardness that was made using a polymer known as PEO. The district court found that the claims were inherently anticipated by McGinity, which discloses a formulation containing PEO. McGinity did not disclose the hardness of the formulation. The Court first reviewed McGinity’s disclosure of active agents, which did not include oxycodone or any opioid analgesic. The Court found that McGinity discloses a formulation containing “a broad group of analgesics,” including oxycodone due to use of terms including “such as” or “and the like.” The district court also relied on, and the Federal Circuit agreed with, evidence showing that the tablets produced according to McGinity had the claimed hardness. In particular, the Court relied on expert testimony regarding the testing of thousands of tablets prepared according to the McGinity disclosure. Purdue failed to produce any evidence that the McGinity tablets would not meet the hardness limitation. Therefore, the Court found that the claim limitation was inherent and that the patent was invalid as anticipated.

It may be valuable to review claims for a possible trigger of product-by-process limitations such as the “derived from 8α” limitation. While infringement of the claims requires meeting the process limitations, the patentability and validity of the claims ignore such limitations. When drafting a product-by-process claim, the focus should be on the novelty or nonobviousness of the product itself. As with highly-pure oxycodone, when the product would have been obvious, the process limitations cannot be used as a

basis for patentability or validity.

While the Court did not cite any prior art disclosing oxycodone with less than 10 ppm of 14-OH, the Court still found that the claims would have been obvious. The FDA requirement showed that there existed an external pressure to lower the amount of 14-OH in oxycodone, regardless of any inventiveness by the patent Applicants. This illustrates that meeting regulatory requirements may be important when evaluating the validity of a patent claim. A regulatory requirement that is the same as a claim element may be used to show that the element would have been obvious.

The fact that the Court found that McGinity discloses oxycodone illustrates that the breadth a disclosure may be greater than what is actually written. It may be useful to evaluate a disclosure based on how it would be read by a person of skill in the art, in order to determine its true scope. The case also shows that when relying on inherency to prove invalidity, reproduction of examples may be very useful. While not every case will require the reproduction of thousands of examples, some degree of testing can help bolster an inherency argument. Expert testimony regarding the nature of products produced by a prior art process can be valuable to prove the inherent extent of the disclosure.