In my earlier post on Ariad v Lilly, there is a detailed discussion of the Federal Circuit’s recent decision that invalidated all of the claims-in-suit for failure to meet the written description requirement. The Federal Circuit rejected Lilly’s attempts to establish inequitable conduct, in part based on the failure of Dr. Baldwin to submit papers he published on the mechanism-of-action of NF-kB after the effective filing date of the ‘516 patent. The panel stated: “Lilly did not show that Dr. Baldwin appreciated the inherent anticipation theory to which the references allegedly pertained.” I would like to spend a little time discussing the inherent anticipation theory and the meaning of the title of this post.

Two requests by third parties (Lilly and Merck?) for reexamination of the ‘516 patent were filed in 2005 and merged as reexamination no. 90/007,503 on May 4, 2006. The reexamination proceedings have continued since then, and the Patent Office has rejected every claim that Ariad has proposed as inherently anticipated by various references. Whether or not the parties will continue this reexamination in view of the verdict, the proceedings make it crystal clear that the Patent Office does not feel it needs to use section 112 to dispose of “mechanism of action” claims such as those that were in suit.

A composite dependant claim 80 was set forth in first my Ariad post. It is a good example of a “Hail Mary” claim, but so is claim 12, which is shorter:

12. A method for reducing the effects of bacterial infection on mammalian immune cells comprised reducing NF-kB activity in mammalian immune cells so as to reduce bacterial [LPS]-mediated stimulation of the immune cells.

Apart from section 112 issues, why do I call this a “Hail Mary claim”? In football, a “Hail Mary” pass is a long throw that the quarterback makes, usually in the last seconds of the game, hoping that one of his receivers, by some miracle, will be there to catch it.

A “Hail Mary claim” is a broad claim that a patent attorney proposes, usually containing broad functional language that, while presumed valid when and if issued, is vulnerable to attack due to non-prior art publications that explain why prior art publications in fact anticipate the claim.

There is no room here to more than begin to summarize the hundreds of pages of rejections, amendments and counterarguments presented by the requestor, patentees’ attorneys, and the Patent Office in the reexamination, but they have stayed focused on this “principle” since the first office action. Put another way, the Patent Office has asserted (and I think that they are correct under the prevailed precedent):

(A) Method claims whose only recited steps are altering biological mechanisms of action are invalid as anticipated, under principles of inherency, by a prior use of a biologically active compound that meets two requirements:
1. More than one year prior to the effective filing date of the claim at issue, an “old compound” was administered so as to treat a condition or achieve a result encompassed by the condition or result recited by the claim at issue; and
2. At any time, it is shown that the mechanism of action of the “old compound”, when used to treat the condition or achieve the result, is the mechanism of action recited in the claim.

To get back to the Ariad reexamination, claims 1 and 12 were found to be inherently anticipated by the 1970 Physicians’ Desk Reference, which provides detailed information about currently marketed drugs, taken with later, non-prior art publications, showing that certain drugs found in the PDR in fact work by inhibiting NF-kB. For, example, the Examiner cited the PDR as teaching the administration of antibiotics such as erythromycin to treat gram negative bacterial infections. The PDR, of course, did not mention NF-kB; it had not been discovered yet. Next, later-generated “intrinsic evidence” was cited to establish that, e.g., erythromycin inhibits LPS and NF-kB activation of cytokine expression (Yasutomi et al., J. Immunol., 175, 8069 (2005)). This is a case where an “inherent species” is anticipating a later generic claim “of vast scope,” to quote the Federal Circuit.

If this seems unfair, consider Eli Lilly & Co. v. Barr Labs., 251 F3d. 955 (Fed. Cir. 2001), where a claim to a method of blocking serotonin uptake by administration of fluoxetine was held to be inherently anticipated by a prior art claim to a method of treating anxiety by administering fluoxetine. The court stated that the later claim “simply describes the process by which fluoxetine hydrochloride physically acts on individuals who receive the drug.” Thus, the later claim does not represent a new use of an “old compound.” Consider that this decision has only been buttressed by more recent decisions relating to inherent anticipation, such as Schering Corp. v. Geneva Pharmaceuticals, Inc., 339 F.3d 1373 (Fed. Cir 2003), which invalidated a claim to a metabolite of an “old drug” because evidence not in existence when the old drug and its use were patented showed that the metabolite formed in the patient’s body when the drug was ingested. This case has been cited by the Examiner throughout the reexamination proceedings for the principle that “inherent anticipation does not require a person of ordinary skill in the art to recognize the inherent disclosure in the prior art at the time the prior art was created.” Citing SmithKline Beecham Corp. v. Apotex Corp., 403 F.3d 1331 (Fed. Cir. 2005), the Examiner has taken the position of the court that “what was actually done, or possible to do, in the prior art was “irrelevant since disclosure, not practice, is necessary for anticipation.” To make the Patent Office’s burden of rejected “Hail Mary” claims even lighter, the Examiner noted that the threshold “for enabling a prior art reference is lower for enablement under [section 112] required for a patented invention insofar that the prior art reference need not demonstrate efficacy or utility.”

Patentee’s attorneys’ counter-arguments have mostly involved the requirement for certainty in the art to support an inherency rejection and that the claims – that have been cancelled or amended to some extent – now contain claim elements not present in the prior art or the intrinsic evidence.

Section 112 issues aside, can the legal prayer of a “Hail Mary” claim ever be answered? I think the answer is a qualified “Yes.” What if a cellular signaling pathway – the mechanism of action – is associated with a condition untreatable with any agent prior to the effective filing date of the claims? For example, consider a claim to a method of halting progression of ALS by administering an agent that inhibits the induction of NF-kB in a mammalian subject. What if a third party discovers an effective “old drug” in 2008 and finds that it works by NF-kB inhibition? Now the claim would not be expressly or inherently anticipated, and there is no need to protect anything in the public domain. This analysis only works if the old drug was not tried prior to the effective filing date of the claim and if no drug that was tried in fact had any efficacy. Perhaps a really thorough search could answer both these questions. Perhaps.