Allergan, Inc. v. Sandoz, Inc.: An Incongruous Obviousness Decision?

This week, the Federal Circuit confirmed that a strong affirmative case of obviousness can overcome unexpected results in Allergan, Inc. v. Sandoz, Inc.  The case involved a number of patents covering Combigan®, a combination eye-drop solution for indicated for glaucoma containing 0.2% brimonidine, an α2-agonist, and 0.5% timolol, a beta-blocker, along with benzalkonium chloride preservative, including a formulation claim and a method claim.  The Federal Circuit found the formulation obvious, but, in seemingly incongruous fashion, that a method of the administering that formulation in a known twice-daily dosing regimen was not obvious.

The Prior Art

The defendants’ obviousness cases rested primarily on DeSantis, which described an ophthalmic combination product containing 0.02 – 2% by weight α2-agonist, 0.01 – 3.0% by weight beta-blocker, and 0.001 – 1.0% by weight “formulatory ingredients,” such as benzalkonium chloride.  DeSantis expressly identified timolol as a suitable beta-blocker for use in the solution, and indirectly identified brimonidine through incorporation by reference to another reference that DeSantis identified as for its disclosure of suitable α2-agonists.  In addition to this evidence, the defendants offered evidence that:

  • Brimonidine was one of only three α2-agonists that the FDA had approved for use in an ophthalmic products at the time;
  • Combination ophthalmic glaucoma products were not new to the scene—the FDA had already approved four of them;
  • Another prior art reference suggested the twice daily, serial administration of 0.2% brimonidine and 0.5% timolol five minutes apart, even though brimonidine monotherapytypically involved administration three times daily; and
  • Other combination products were administered twice daily even though one of the agents in the combination required dosing three times daily.

The Claimed Combination Formulation was Obvious

Despite these prior art teachings, the district court found that a claim directed to a combination of 0.2% brimonidine and 0.5% timolol was not obvious.  The court based this decision in part on evidence that prior attempts to administer brimonidine only twice daily had failed, whereas the claimed combination maintained efficacy despite administration only twice daily.  The Federal Circuit reversed.  On the issue of unexpected results, the Federal Circuit accepted the district court’s finding that the twice daily dosing was an unexpected benefit of the combination product, but that this advantage was insufficient to overcome the clear motivation to develop a combination shown in the prior art literature.  In reaching its decision, the Federal Circuit also held:

  • Improved patient compliance is sufficient motivation, even if improved compliance is not among the criteria considered by the FDA when deciding to whether to approve a new drug product;
  • A subtle teaching away from a claimed modification of the prior art may be overcome by a strong motivation to make such a modification;
  • While formulation science does carry a level of unpredictability, some risk of failure can exist without defeating the obviousness of an invention; and
  • Difficulty developing the branded product / commercial embodiment is relevant only insofar as such difficulties relate to the claimed elements of the invention, and not unclaimed aspect of the product.

The Method Claim Was Not Obvious

Despite finding the combination of the drugs into a single composition was obvious (with one motivation being to improve patient compliance), the Federal Circuit ruled that a method of administering that combination twice daily was not obvious.  “[The defendant] has failed to point to evidence that would allow us to conclude that the addition of timolol to brimonidinedosed twice per day would eliminate [the loss of efficacy in the afternoon that occurred with twice daily brimonidinemonotherapy].”  The Court went on to state that “while it is true that the prior art shows [twice daily administration of the claimed combination], this art does not show that there was no loss of efficacy associated with that treatment, let alone elimination of the afternoon [efficacy problem with twice daily administration of brimonidine.]”  It appears the Court was confused by the fact that the alleged inventors had crafted their method claim to expressly reference the “unexpected result.”

Judge Dyk dissented based on the well-established principle that “newly discovery results of known processes directed to the same purpose are not patentable.”  As Judge Dyk aptly noted, “avoiding a ‘loss of efficacy’ [in the afternoon] is not a separate step, but rather a result of the claimed method.”

It seems to us quite clear that even when an unexpected result is written into the wording of a claim, it should not render an otherwise obvious method patentable.  Methods should be defined, and their patentability assessed, by the claimed steps—not the results of the method.  The majority attempts to sidestep this issue by suggesting the defendants had failed to demonstrate that the unexpected result was inherent to the otherwise obvious method.  But this position seems equally misguided, because if the unexpected (and claimed) result is not inherent to the claimed method, how is the claim enabled.  Do we agree on this one?