Pharmas Walking the Patent Tightrope

Includes: LLY, TEVA
by: Aaron F. Barkoff

Janssen Pharmaceutica N.V. v. Teva Pharms. (NYSE:TEVA) USA, Inc. (Fed. Cir. 2009)

Eli Lilly & Co. (NYSE:LLY) v. Actavis Elizabeth LLC, et al. (D.N.J. 2009)

Two recent cases illustrate how owners of pharmaceutical method-of-use patents may find themselves walking a tightrope when their patents are attacked simultaneously on enablement and obviousness grounds.

First, in a 2–1 decision (.pdf) last September, the Federal Circuit affirmed the invalidity of U.S. Patent No. 4,663,318. The ‘318 patent claims “a method of treating Alzheimer’s disease and related dementias which comprises administering . . . a therapeutically effective amount of glanthamine. . . .” The ‘318 patent is the only Orange Book-listed patent for Janssen’s Alzheimer’s drug, Razadyne.

Then, in an opinion (.pdf) filed in December, denying motions for summary judgment, the U.S. District Court in New Jersey clearly indicated that Eli Lilly would have a hard time at trial surviving the defendants’ lack-of-enablement and obviousness challenges to U.S. Patent No. 5,658,590. The ‘590 patent claims a method of treating Attention Deficit/Hyperactivity Disorder (ADHD) with atomoxetine. The ‘590 patent is the only Orange Book-listed patent for Lilly’s ADHD drug, Strattera.

In both cases, the generic drug company defendants asserted that the patents were invalid for both lack of enablement and obviousness. More specifically, they argued that the claimed methods were not enabled because the patentees had not properly established that the claimed methods had utility.

The Federal Circuit described the relationship between the enablement and utility requirements in Process Control v. HydReclaim:

The enablement requirement of 35 U.S.C. 112, requires that the specification adequately discloses to one skilled in the relevant art how to make, or in the case of a process, how to carry out, the claimed invention without undue experimentation. The utility requirement of 35 U.S.C. 101 mandates that any patentable invention be useful and, accordingly, the subject matter of the claim must be operable. If a patent claim fails to meet the utility requirement because it is not useful or operative, then it also fails to meet the how-to-use aspect of the enablement requirement.

In Janssen, the Federal Circuit further explained:

The utility requirement prevents mere ideas from being patented. . . . The utility requirement also prevents the patenting of a mere research proposal or an invention that is simply an object of research.

Quoting a 1966 Supreme Court decision, Brenner v. Manson, the Janssen court explained the policy behind the utility requirement:

Allowing ideas, research proposals, or objects only of research to be patented has the potential to give priority to the wrong party and to confer power to block off whole areas of scientific development, without compensating benefit to the public.

According to the Janssen decision, “Typically, patent applications claiming new methods of treatment are supported by test results.” But “human trials are not required for a therapeutic invention to be patentable, as results from animal tests or in vitro experiments may be sufficient to satisfy the utility requirement.” Further, utility/enablement is determined as of the effective filing date of the patent application and it must be recognizable to one of ordinary skill in the art from the written description. Therefore, any test results (whether human, animal or in vitro) should be disclosed in the original patent application. The patentees in Janssen and Lilly ultimately obtained test results supporting the utility of their claimed methods of use, but they did not include those test results in their patent applications.

Thus, Janssen and Lilly argued that one of skill in the art would have recognized the utility of their inventions from the discussion of prior art in their patent specifications. For example, Janssen argued that (1) the prior art tests summarized in the specification would lead one skilled in the art to infer that galantamine affected the ability of acetylcholine to bind to both nicotinic and muscarinic receptors in the brain; (2) the animal tests proposed in the specification as a model for Alzheimer’s disease would further lead one skilled in the art to infer that the model’s method of impairing brain acetylcholine availability would allow both muscarinic and nicotinic effects to be observed; and (3) because nicotinic receptors in the brain are involved with the ability to learn, the specification suggested that galantamine could have beneficial effects on learning. Similarly, Lilly argued that one of ordinary skill in the art would have inferred that: (1) the prior art compounds disclosed in the specification were effective in treating ADHD because they inhibited norepinephrine; (2) atomoxetine was, similarly, an active norepinephrine inhibitor; and (3) atomoxetine would therefore be successful in treating ADHD.

And this is where Janssen and Lilly’s non-obviousness arguments caused them trouble.

For instance, in Janssen, the Federal Circuit majority relied upon the following testimony (all of which was provided in support of non-obviousness) to affirm the finding of no enablement:

  • During prosecution, in response to the examiner’s obviousness rejection, the inventor stated that “nothing in this teaching leads to an expectation of utility against Alzheimer’s disease.”
  • The inventor also stated that “predicting that galanthamine would be useful in treating Alzheimer’s disease just because it has been reported [in the prior art studies cited in the specification] to have an effect on memory” would be baseless.
  • At trial, Janssen’s expert testified that studying a compound’s effects on scopolamine-induced amnesia [as disclosed in prior art cited in the specification] “ignores the whole other [nicotinic] part that’s damaged in Alzheimer’s disease” and thus “doesn’t mimic Alzheimer’s disease.”
  • Also at trial, the inventor testified, “When I submitted this patent, I certainly wasn’t sure, and a lot of other people weren’t sure that cholinesterase inhibitors, [a category of agents that includes galantamine,] would ever work.”

In Lilly, the district court relied upon similar testimony:

  • The inventor admitted that treating ADHD with atomoxetine was merely hypothetical.
  • Lilly’s expert testified that if in 1995 he heard the hypothesis that atomoxetine would be effective in treating ADHD, he would have responded: “you’re wrong, it’s not a good model.”
  • Lilly’s expert also explained that a person of ordinary skill in the art “would not predict atomoxetine would be effective” as “we were moving away from single neurotransmitter drugs.”

A footnote in the Lilly case nicely summarized the problem that a dual lack-of-enablement/obviousness attack presents to patentees:

Plaintiff emphasizes the differences between atomoxetine and the prior art for the purposes of refuting Defendant’s obviousness argument, while at the same time asserting that the prior art and atomoxetine are in some ways similar in order to demonstrate enablement/utility. Defendants argue, then, that the Court must find the patent invalid as either obvious or not enabled. For example, if the Court determines that a person of ordinary skill in the art would be able to infer utility based upon the patent’s specification, Defendants’ enablement argument might fail, but its obviousness argument would presumably be bolstered. In essence, Defendants argue that whichever set of experts is credited, Plaintiff’s patent will be invalidated.

In the Janssen and Lilly cases, perhaps the only way to avoid this Catch-22 would have been to include preliminary test data in the orignal patent specification.