The Federal Circuit showed once again the importance of a district court's factual findings (and the deference the appellate court gives those findings, particularly when supported by expert testimony), in InSite Vision Inc. v. Sandoz, Inc.  In last week's decision in this case the Court affirmed a finding of non-obviousness of four Orange Book-listed patents challenged by generic drugmaker Sandoz in ANDA litigation under 35 U.S.C. § 271(e)(2).

The case involved a topical formulation of the antibacterial drug azithromycin formulated for ocular administration for the treatment of bacterial infections in the eye and sold by Insite as Azasite®.  The four patents-in-suit were U.S. Patents Nos. 6,861,411 (the "'411 patent"); 6,239,113 (the "'113 patent"); 6,569,443 (the "'443 patent"); and 7,056,893 (the "'893 patent"); the first patent was owned by Pfizer and licensed to InSite and the rest were owned solely by InSite.  Representative claims at issue are as follows:

Claim 1 of the '411 patent:

1.  A method of treating an ocular infection, comprising topically administering to an eye of an animal in need of such treatment an ocular infection-treating amount of azithromycin.

Claim 1 of the '113 patent:

1.  A process for treating an eye, which comprises: topically applying an aqueous polymeric suspension of an azalide antibiotic, wherein said suspension comprises water, 0.01% to 1.0% of an azalide antibiotic, and 0.1 to 10% of a polymeric suspending agent.

Claim 16 of the '443 patent:

16.  A topical ophthalmic composition comprising an aqueous polymeric suspension comprising water, 0.01% to 1.0% of an azalide antibiotic and 0.1 to 10% of a polymeric suspending agent, wherein said topical ophthalmic composition has an osmotic pressure of from 10 to 400 mOsM and wherein said composition does not contain constituents that are physiologically or ophthalmically harmful to the eye.

Claim 1 of the '893 patent:

1.  A composition comprising water, a polymeric suspending agent and an azalide antibiotic, wherein said composition has a pH of about 6.0 to 6.6.

After claim construction Sandoz conceded infringement but challenged the asserted claims on obviousness grounds.  The District Court found that Sandoz had failed to show by clear and convincing evidence that the claims of the Orange Book-listed patents were invalid for obviousness, and Sandoz appealed.

The Federal Circuit affirmed, in an opinion by Opinion by Judge Linn, joined by Chief Judge Prost and Judge Newman.  The panel first addressed the District Court's non-obviousness determination regarding the '411 patent, specifically with regard to the question of how the obviousness issue was framed.  InSite argued that the District Court had properly considered whether it would have been obvious to the skilled worker to produce a topical azithromycin formulation for ophthalmic uses, while Sandoz contended that the proper question was whether it would have been obvious to use azithromycin topically to treat conjunctivitis.  These considerations impacted Sandoz' burden of proof: framed narrowly, which Sandoz contended was proper, they would not have the burden of showing "the obviousness of topical treatments of all manner of eye infections and not merely conjunctivitis with azithromycin."  The District Court agreed with InSite that there was "no reason to limit the question to conjunctivitis and to azithromycin," based on the need in the art for such formulations for treating things other than conjunctivitis, including corneal infections.

The Federal Circuit noted in its opinion that this question of "framing" the obviousness issue was one of fact, and thus that on appeal Sandoz had the burden of showing clear error in the District Court's decision not to narrow the scope of the obviousness question to treatment of conjunctivitis with azithromycin.  On that basis, the panel found no clear error, at least in part because the Court opined that nothing stopped Sandoz from presenting evidence that it would have been obvious to use the claimed formulations to treat conjunctivitis (which, the panel apparently believed, could have resulted in Sandoz prevailing).  The Federal Circuit panel held that the District Court "did not clearly err" on this record.  Citing In re Kahn, the Court noted that motivation sufficient to establish obviousness is not limited to the 'specific problem solved by the invention," 441 F.3d 977, 988 (Fed. Cir. 2006), and that "[d]efining the problem in terms of its solution reveals improper hindsight in the selection of the prior art relevant to obviousness," citing Monarch Knitting Mach. Corp. v. Sulzer Morat GmbH, 139 F.3d 877, 881 (Fed. Cir. 1998).  Sandoz' argument, that the District Court should have defined the "problem" addressed more narrowly, could itself be reversible error, according to the Court, because "an overly narrow 'statement of the problem [can] represent[] a form of prohibited reliance on hindsight, [because] [o]ften the inventive contribution lies in defining the problem in a new revelatory way,'" citing  Mintz v. Dietz& Watson, Inc., 679 F.3d 1372, 1377 (Fed. Cir. 2012).

The panel recognized that Sandoz' primary authority, Alcon Research, Ltd. v. Apotex Inc., 687 F.3d 1362, 1368 (Fed. Cir. 2012), stands for the proposition that "if the prior art would motivate a person of skill in the art to make the claimed invention, even if that was not based on 'the same motivation that the patentee had,' the patent would have been obvious" (emphasis in opinion).  However, the opinion noted that this is a question of fact, and that the District Court did not err in deciding that the problem in the art addressed by the claimed invention was broader than preparing topical azithromycin formulations for treating conjunctivitis.  Specifically, the District Court found, and the panel approved, that azithromycin was known in the art to be a "poor choice" for topical administration to the eye, based on its intrinsic chemical properties ("the unique balance of log P, molecular weight, solubility, and charge").  The panel also noted that the District Court relied upon and credited the testimony of InSite's expert regarding the understanding in the art that the ability to penetrate the cornea and not just the conjunctiva was important to avoid spread of a conjunctival infection into the cornea.

The panel summarized its conclusions regarding Sandoz' argument on whether the District Court properly "framed" the obviousness issue by stating:

[T]he district court's framing of the question did not prevent Sandoz from attempting to invalidate the asserted claims, which were not limited to the cure of conjunctivitis, by proving that it would have been obvious to use azithromycin in a topical treatment to cure that one infection.  The problem for Sandoz, as we will next address, is that its proofs simply failed to carry the day in satisfying its clear and convincing burden.

Turning to the merits, the panel characterized Sandoz' argument to be that the claimed formulation was an obvious variant of a prior art formulations for topical administration of erythromycin and oral administration of azithromycin sold as Zithromax®.  Based on the testimony of its expert, Sandoz contended that substituting azithromycin for erythromycin in topical formulations for treating conjunctivitis would have been obvious, due to the similarities between the two antibacterial drugs.  Another Sandoz expert testified that making a topical formulation of azithromycin would have been obvious because it was known in the art that "topical treatments [were] more effective than oral treatments." Sandoz also relied upon purported public suggestions for using azithromycin in topical formulations for treating conjunctivitis.

The panel was not persuaded by Sandoz' arguments, based on its judgment that "[t]he district court thoroughly and properly considered all of the evidence presented and the various arguments raised by the parties in ruling the asserted claims to be not invalid."  This evidence included the existence of preferred alternative ophthalmic treatments, particularly ones using fluoroquinolones, which the Court determined were known in the art to be a better choice than azithromycin, due to their known antibacterial properties and better penetration of ocular tissues like the conjunctiva and cornea.  These properties were in contrast to azithromycin, which the skilled worker would have had reason to anticipate would not be as able to penetrate such ocular tissues (due to, inter alia, its higher molecular weight).  The panel also found no error in the District Court crediting InSite's expert over Sandoz' expert's testimony, and in discounting the relevance of prior art formulations in view of differences of opinion by the experts (including evidence that Sandoz' expert had his own patent for topical ophthalmic treatments that included "24 potential antibiotics, including erythromycin but did not list azithromycin").

The panel also affirmed the District Court's non-obviousness determination for the remaining formulation patents (the '113 patent, '443 patent, and '893 patent) owned by InSite.  Here, the Court based its opinion on the District Court's finding that the skilled worker would not have been motivated to use "water-based polymeric solutions" known in the prior art when formulating azithromycin because the drug was "considered insoluble and unstable in water."  The District Court also found that the skilled worker would have used a colloidal system, rather than the claimed gelling polymer, in view of the teachings in the prior art.  The panel rejected Sandoz' contention that a prior InSite patent, U.S. Patent No. 5,192,535, taught erythromycin formulations having the same gelling polymer recited in InSite's claims, and that this teaching would have made it obvious to substitute azithromycin for erythromycin to arrive at the claimed formulation.  The Federal Circuit agreed with InSite and the District Court that the '535 patent taught "a laundry list of active ingredients" and that that patent did not exemplify formulations of such gelling polymers with erythromycin (again, based on expert testimony).  In reaching this conclusion the panel distinguished the teachings of Merck & Co., Inc. v. Biocraft Labs., Inc., 874 F.2d 804, 807 (Fed. Cir. 1989), asserted by Sandoz in support of its obviousness arguments, on the basis that in Merck the prior art "expressly taught the combination of compounds claimed in the [later] patent," which was not the case here.  And it was also significant that not only did the '535 patent not expressly teach the combination of erythromycin with a gelling polymer but that the skilled worker would have to modify the patent's teachings to arrive at the invention claimed in the patents in suit.

Finally, the opinion said there was no clear error in the District Court's finding that there were "meaningful" secondary considerations of non-obviousness (although in view of the remainder of the opinion it is clear that the Federal Circuit did not need to rely on these factors).

In sum, the opinion shows that the Federal Circuit will defer to district court factual determinations in questions involving ultimate questions of law, such as non-obviousness and that an apparently important factor in deciding to defer is a district court's reliance on expert testimony to resolve factual disagreements between the parties.  While this behavior is consistent with how courts traditionally parse review by appellate courts, compliance with these traditional standards reintroduces some uncertainty into the application of patent law doctrines that, while inevitable because the facts of each case are necessarily different, is a far cry from the consistency in outcome that the Court spent its first quarter century trying to erect.  Whether this was a prudent or even practical course is moot:  more than a decade of Supreme Court disapproval has clearly influenced the Federal Circuit to adopt the presumptions and practices of its Sister Circuits in such matters.

InSite Vision Inc. v. Sandoz, Inc. (Fed. Cir. 2015)
Panel: Chief Judge Prost and Circuit Judges Newman and Linn
Opinion by Circuit Judge Linn