Case Nos. 2011-1619 et al. (Dyk, Prost, O’Malley)
When nonobviousness is based solely on a property or functionality, should patentability differ if the invention is recited as a composition having the property versus as a method exploiting the property? Yes, says a majority panel of the Federal Circuit in Allergan v. Sandoz. An interesting result, and the way the court arrives there offers some good lessons on the flexible, and sometimes complex, nature of the modern § 103 inquiry.
Sandoz and several others sought ANDA approval to market a generic version of Allergan’s Combigan, covered by US Patents 7,642,258; 7,320,976; 7,323,463; and 7,030,149. The patents described an eyedrop combining 0.2% brimonidine and 0.5% timolol for treating glaucoma. The district court found all relevant claims of the patents valid, including these formulation and administration claims:
1. A composition comprising about 0.2% timolol by weight and about 0.5% brimonidine by weight as the sole active agents, in a single composition.
4. A method of reducing the number of daily topical ophthalmic doses of brimondine administered topically to an eye of a person in need thereof for the treatment of glaucoma or ocular hypertension from 3 to 2 times a day without loss of efficacy, wherein the concentration of brimonidine is 0.2% by weight, said method comprising administering said 0.2% brimonidine by weight and 0.5% timolol by weight in a single composition.
The district court found these claims not invalid – nonobvious – over prior art of US Patent 5,502,052 to DeSantis in view of papers by Timmermans, Larsson, and some ophthalmology journal articles. The content of this prior art differed minimally from the scope of the claims. DeSantis disclosed a combination therapy of timolol (a beta blocker) with alpha2-agonists (a very small class to which brimonidine belonged – Timmermans, incorporated by reference in DeSantis, specifically disclosed brimonidine as such) in the claimed concentrations. While DeSantis did not state that the two drugs were mixed in a single formulation, the patent did strongly suggest their use together and observed difficulties in patient compliance in having to administer multiple drops per day. The Larsson article disclosed separate administrations of 0.2% brimonidine and 0.5% timolol spaced 5 minutes apart for treatment of eye problems. Several other prior art articles and products used combined formulations for single administrations containing timolol, all showing increased efficacy in treating glaucoma.
The district court gave more credit to evidence of non-obviousness over the closeness of the prior art. Particularly, FDA rules gave no incentive to combine therapies into a single formulation because patient compliance is not a factor in FDA approval; opthalmolic formulations were rather unpredictable; other articles taught away from using a combined therapy; and there were secondary considerations of long-felt need and unexpected results.
On appeal, the Federal Circuit dismissed each of these elements of nonobviousness in light of the strong prima facie case of invalidity as to the formulation claims. For motivation to combine, the court noted that while regulatory approval can be considered in the flexible obviousness inquiry, motivation to combine can come from anywhere. Simply because the FDA approval process did not incent the claimed combination therapy did not undo the fact that DeSantis recommended the combination:
The potential for FDA approval also may properly be considered, as it was here, in determining whether one of ordinary skill would be motivated to develop a drug product and whether there was skepticism regarding the efficacy of such a product. Nevertheless, we find the district court erred in concluding that one of ordinary skill would not be motivated to develop a fixed combination product to increase patient compliance because the FDA did not consider that particular motivation when evaluating drug applications. There is no requirement in patent law that the person of ordinary skill be motivated to develop the claimed invention based on a rationale that forms the basis for FDA approval. Motivation to combine may be found in many different places and forms; it cannot be limited to those reasons the FDA sees fit to consider in approving drug applications.
When viewed under the proper standard, the evidence of record establishes a motivation to combine brimonidine and timolol into a fixed combination product. Not only does DeSantis teach the fixed combination of timolol with an alpha2-agonist, numerous other references teach the fixed combination of other ophthalmic drugs. Allergan, Inc., 818 F. Supp. 2d at 1016-17. In fact, there were at least four other fixed combination products for the treatment of ocular hypertension and glaucoma on the market at the time of invention. J.A. 631. Moreover, it was common at the time of the invention to provide brimonidine and timolol to a patient in serial fashion and DeSantis taught that by combining drugs in a fixed-combination formulation, patient compliance could be increased. Accordingly, we find clear error in the district court’s finding that there was no motivation to develop a fixed combination brimonide/timolol product
Similarly, on unpredictability, the court noted that the single therapy was essentially shown to likely work from DeSantis. This reasonable expectation of success was enough to counter nonobviousness: “While we agree that formulation science carries with it a degree of unpredictability, ‘obviousness cannot be avoided simply by a showing of some degree of unpredictability in the art so long as there was a reasonable probability of success.’ Pfizer, Inc. v. Apotex.” The court was also unmoved by difficulties Allergan had in formulating the commercial product Combigan, because that product included problematic ingredients not recited in the claims.
On teaching away and secondary factors, the court agreed they were present to some extent (expect for long-felt need, which the court dismissed as conclusory) but could not overcome the strong prima facie showing of obviousness based on the closeness of the prior art. Although the synergy between the claimed brimonidine and timolol was real – it had a much longer-lasting effective dose – a person skilled in the art still had other motivation to create the claimed formulation regardless of this bonus:
previous attempts to treat patients twice per day with brimonidine resulted in a loss of efficacy eight to nine hours post administration. This loss of efficacy is referred to as the “afternoon trough.” The court found that a twice per day dosage regimen of Combigan® unexpectedly did not suffer from the afternoon trough issue. We agree with the court’s finding that this result was unexpected. However, we do not find that these unexpected results are sufficient to outweigh the other evidence of obviousness as to these formulation claims. While the unexpected benefits of twice a day dosing of the combination formula are relevant to Sandoz’s attack on the validity of the method claims, we do not find it similarly meaningful to our analysis of the formulation claims. There is extensive evidence in the prior art showing the concomitant administration of brimonidine and timolol multiple times per day, that the combination had benefits over the administration of either alone, and that there was a motivation to combine the two to achieve better patient compliance. KSR, 550 U.S. at 426. Whether or not that combination also solved problems associated with the afternoon trough, we find the motivation to make the combination was real.
And here the majority and dissent parted ways. As seen in the above excerpt, the majority started hedging about the unexpected synergy being differently applicable to the formulation and method claims. The majority found the unexpected results of fewer doses required by a timolol + brimonidine fixed combination product was not apparent from the prior art. This increased efficacy was especially applicable to the method claims specifically reciting the dose reduction, and the efficacy might not have been observed in every type of administration:
[footnote 1] The evidence of record does not establish that the dose reduction “from 3 to 2 times a day without loss of efficacy” limitation is an inherent property or a necessary result of the administration of 0.2% brimonidine and 0.5% timolol in a single composition. Of course, it may be true that the mere administration of 0.2% brimonidine and 0.5% timolol twice daily in any fixed combination formulation inherently produces the claimed result. Alternatively, it may also be true that only certain fixed combination formulations produce this result. On the present record, we cannot draw a conclusion in favor of either proposition.
For the dissent, if the formulation having the unexpected property was nonetheless obvious, the simple method of exploiting the property was similarly obvious:
In this case, the method of claim 4 consists of a single step: applying a fixed combination of 0.2% brimonidine and 0.5% timolol twice a day. See ’149 patent col. 10 ll. 10- 17. This method was surely obvious to try. See KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 421 (2007).
. . .The majority’s outcome appears to rest, therefore, on the notion that claim 4 was not obvious because it claims the result of twice-a-day dosing—avoiding “a loss of efficacy in the afternoon.” See Maj. Op. 13. Avoiding a “loss of efficacy” is not a separate step, but rather a result of the claimed method. See Bristol-Myers Squibb, 246 F.3d at 1374-78; see also Abbott Labs., 471 F.3d at 1369. We should recognize in this case, as we did in Bristol-Myers Squibb, that “[n]ewly discovered results of known processes directed to the same purpose are not patentable.” Bristol-Myers Squibb, 246 F.3d at 1376.
[footnote 1] The majority . . . suggests that this rule should not apply here because there may exist specific formulations of a fixed combination of 0.2% brimonidine and 0.5% timolol that do not inherently achieve this result. See Maj. Op. 14 n.1. Claim 4, however, is not limited to any particular formulation. See ’149 patent col. 10 ll. 10-17. The majority’s argument therefore only suggests that the claim would have been even more clearly obvious, since it would cover the use of compositions that do not even achieve the allegedly unexpected result. “Claims [that] are broad enough to read on obvious subject matter are unpatentable even though they also read on nonobvious subject matter.” In re Lintner, 458 F.2d 1013, 1015 (CCPA 1972); see also ArcelorMittal Fr. v. AK Steel Corp., 700 F.3d 1314, 1325 (Fed. Cir. 2012) (citing Lintner); Muniauction, Inc. v. Thomson Corp., 532 F.3d 1318, 1328 n.4 (Fed. Cir. 2008) (same).
Summary judgment of nonobviousness reversed, except as to the method claim 4.
Lessons: As secondary considerations go, unexpected results – a new or improved property or function in a claimed combination – are among the trickiest to leverage. The results have to be impressive, they have to be absent – even inherently – from the prior art, they have to exist substantially throughout the whole claim scope, and they typically need to be shown through reliable expert evidence. Moreover, while effective in rebutting weaker prima facie obviousness showings, unexpected results seem to lose steam fastest among the secondary considerations as the prima facie obviousness case strengthens. Allergan showcased a very strong prima facie § 103 rejection with little distance between prior art and claims and motivation to close that gap independent of any newfound benefits in the combination, and the court was fine dismissing teaching away and unexpected results against such a case. It appears prudent to fairly assess any prima facie case brought against claims in prosecution and litigation; if the case is strong, with good reasoning to otherwise arrive at the claimed structure, be especially wary of bringing only unexpected results to the table.
If Allergan’s majority is widely followed, it counsels for inclusion of method claims that specifically recite and use an unexpected result. Consider drafting such claims if the unexpected results are given in the specification or adding them in prosecution when unexpected results are proffered via declaration in prosecution. As the dissent makes clear, this is still not fool-proof, especially for rather bland method claims. Indeed, we’ve seen the Federal Circuit recently dismiss unexpected results as inherent in simple methods that implement an otherwise obvious structure before in In re Taylor and In re Kao. (But compare Unigene, where a newly recognized property in an otherwise obvious formulation saved the formulation claims from obviousness even though the property was never recited. I don’t recommend trying that at home.)
One might consider reciting the unexpected results, i.e., the property or functionality, in the composition claim as well. The patentee in Allergan did not limit the formulation claims to those with increased efficacy, and the majority in part identified that failure in their reasoning invalidating the formulation claims. But I’m not convinced this will get anywhere beyond a method claim as recommended in the previous paragraph: In re Mousa shows that panels are willing to disregard recited properties or functionalities in non-method claims up against a strong prima facie showing of obviousness.
Comment: I think the dissent has the better position here. What it comes down to for me is the dissent’s observation that the method – administering the combination therapy – was the obvious (to try) action to take with the formulation. This should have required the same obviousness determination between the formulation and method of use. If a claimed structure is obvious despite an unexpected result, any obvious use of the structure that achieves only the unexpected result should be obvious too. Otherwise you withdraw a structure from the public domain exactly for what was apparently so unimpressive to keep the structure in the public domain. Of course, if the method is more nonobvious than simply “use [x] to achieve [unexpected result],” bifurcated obviousness between obvious structure and new method of using the structure makes sense and mirrors existing § 103 jurisprudence allowing claims to new methods of using old structures. Where the majority and dissent sparred in footnotes on this – the differing scope between the formulation and method claims – I think the dissent was better positioned in finding that the method claims had broader, and thus obvious, scope.