By Kevin E. Noonan --
Last Thursday, the Federal Circuit handed down its non-precedential decision in Purdue Pharma v. Depomed, reviewing the decision of the Patent Trial and Appeal Board on three related inter partes reviews. While not quite a unicorn, the PTAB decision was unusual in that it affirmed the validity of all claims at issue over a challenge for obviousness.
The challenged patents, U.S. Patent Nos. 6,340,475 and 6,635,280, claimed "controlled-release oral dosage form[s] of a soluble drug and a method of use thereof," which comprised drug dispersed within a solid polymeric matrix. Such dosage forms were structured so that, after administration, the matrix would swell in the stomach (from absorbing water) and as a consequence remain in the stomach in the presence of food. As a result, release of the drug from the matrix and adsorption by the body is delayed, which reduces the risk of overdose and controls drug dosage over an extended period of time at safer and more effective levels.
The three IPRs were directed to the following claims, all asserted to be invalid for obviousness: (1) claims 1, 8–10, 13–15, 43, 45, and 46 of the '280 patent; (2) claims 1, 8–10, 13–15, 61, and 62 of the '475 patent; and (3) claims 43, 54, 55, 57, 58, and 66 of the '475 patent. Claims 1 and 43 of the '475 patent are representative:
1. A controlled-release oral drug dosage form for releasing a drug whose solubility in water is greater than one part by weight of said drug in ten parts by weight of water,
said dosage form comprising a solid polymeric matrix with said drug dispersed therein at a weight ratio of drug to polymer of from about 15:85 to about 80:20,
said polymeric matrix being one that swells upon imbibition of water thereby attaining a size large enough to promote retention in the stomach during said fed mode ["the swelling limitation"],
that releases said drug into gastric fluid by the dissolution and diffusion of said drug out of said matrix by said gastric fluid,
that upon immersion in gastric fluid retains at least about 40% of said drug one hour after such immersion and releases substantially all of said drug within about eight hours after such immersion,
and that remains substantially intact until all of said drug is released ["the substantially intact limitation"].
43. A method of administering to a subject a drug that is therapeutic to said subject when absorbed in the stomach where said drug has at least one ionized group in the pH range 5 through 8,
said method comprising orally administering to said subject a dosage form of said drug while said subject is in a fed mode,
said dosage form comprising a solid polymeric matrix with said drug dispersed therein at a weight ratio of drug to polymer of from about 0.01:99.99 to about 80:20,
said polymeric matrix being one that:
(a) swells upon imbibition of gastric fluid to a size large enough to promote retention in the stomach during said fed mode ["the swelling limitation"],
(b) releases said drug into gastric fluid by the dissolving of said drug by said gastric fluid and either erosion of said matrix or diffusion of said dissolved drug out of said matrix,
(c) retains at least about 40% of said drug one hour after such immersion in gastric fluid,
(d) releases substantially all of said drug within about ten hours after such immersion, and
(e) remains substantially intact until all of said drug is released ["the substantially intact limitation"],
thereby extending the release rate of said drug with time during said fed mode while releasing substantially all of said drug within said stomach where said drug is maintained in an acidic environment.
Petitioner Purdue Pharma asserted several prior art references, only two of which were relevant to the PTAB's decision and reviewed by the Federal Circuit: a scientific journal article by Baveja et al., Zero-Order Release Hydrophilic Matrix Tablets of β-Adrenergic Blockers, 39 Int'l J. Pharmaceutics 39 (1987) ("Baveja"), and U.S. Patent No. 5,582,837 ("Shell").
The PTAB held that Purdue had failed to establish that the claims were invalid for obviousness by a preponderance of the evidence, the Board standard in IPRs. It agreed with Purdue that the Baveja reference disclosed delayed release drug formulations but failed to disclose the "swelling" and "substantially intact" limitations, and that the Shell patent disclosed drug formulations comprising these missing limitations. Nevertheless, the Board held that Purdue failed to show that the art provided "a reasoned basis for combining these references with a reasonable expectation of success" at achieving the claimed formulations. Specifically, the Board held that Purdue had failed to show the "how or why" the skilled worker would have combined the features from the Shell patent in the Baveja formulation. In addition, the PTAB held that insofar as the motivation to combine the cited references was determined by the problem to be solved it constituted hindsight reconstruction and thus was not a proper basis for finding the claims to be obvious.
In addition, the Board found that Purdue had not established that there would have been a reasonable expectation by the skilled worker of success, based on expert testimony proffered by the patentee on the number of variables involved in combination with inventor testimony regarding how long it had taken to achieve the claimed formulation. In addition, the Board held that Purdue had not shown that adding the swelling and substantially intact features would not change the other properties of the formulation.
The Federal Circuit affirmed, in an opinion by Judge Lourie, joined by Chief Judge Prost and Judge Newman. The Court reviewed the Board's legal conclusion of obviousness de novo and its factual determinations for substantial evidence (citing In re Gartside). Moreover, concerning the questions before it, the panel stated that "the determinations of what a reference teaches and the existence of a reason to combine references are questions of fact," citing In re Beattie, 974 F.2d 1309, 1311 (Fed. Cir. 1992), and In re Hyon, 679 F.3d 1363, 1365–66 (Fed. Cir. 2012). The opinion agreed that there was substantial evidence supporting the Board's determination that "each limitation of the challenged independent claims was known in the art, as evidenced by the teachings of Baveja and Shell." However, the Court also agreed that Purdue bore the burden of showing that the skilled worker had a reason to combine these teachings and have had a reasonable expectation of success, citing PAR Pharm., Inc. v. TWI Pharm., Inc., 773 F.3d 1186, 1193 (Fed. Cir. 2014). The Court credited the expert opinion relied upon by the Board that the teachings of the cited references were "interrelated," but found that Purdue "did not explain in sufficient detail how or why a skilled artisan would have been motivated to combine the 'swelling' and 'substantially intact' features of the Shell formulation with the Baveja formulation to attain the claimed dosage form." Regarding the Board's hindsight determination, the Court found that "[Purdue] failed to demonstrate to the Board that the problem was known in the art or that Purdue's formulation of the problem was derived directly from the prior art, rather than from the challenged claims," citing Insite Vision Inc. v. Sandoz, Inc., 783 F.3d 853, 859 (Fed. Cir. 2015), for the proposition that "[d]efining the problem in terms of its solution reveals improper hindsight in the selection of the prior art relevant to obviousness." The panel also agreed with the Board that Purdue had failed to show that the skilled worker would have expected that modifying the dosage formulation disclosed by Baveja to incorporate the features of the Shell patent would not affect (presumably negatively) the "other desired properties" of the patented formulation.
While being of limited significance (constrained by its facts and lacking precedential value), the case may provide some comfort that the PTAB can in the right circumstances uphold the validity of a pharmaceutical claim challenged for obviousness with sufficient legal support to be affirmed by the Federal Circuit.
Purdue Pharma L.P. v. Depomed, Inc. (Fed. Cir. 2016)
Panel: Chief Judge Prost and Circuit Judges Newman and Lourie
Opinion by Circuit Judge Lourie