Docs at BIO: "Gotcha" Games Continue at USPTO

    By Kevin E. Noonan --

As reported by Patent Docs earlier this week, the U.S. Patent and Trademark Office took the podium at BIO 2008 last week in San Diego, in the person of Dr. George Elliott, a Director in Group 1600.  Like his boss, John Doll, Director Elliott is a personable and friendly PTO presence as a speaker, and he did a good job of explaining Patent Office initiatives like worksharing programs and how the Group is applying the KSR Int'l Co. v. Teleflex Inc. decision.  It was when he turned to his last topic, the "new" written description training materials (to be used by examiners in implementing the "old" written description guidelines promulgated in 2001) that his talk reminded us that we were listening to a PTO official from the Dudas administration.

Specifically in this regard, Dr. Elliott spoke about Example 11, dealing with how the written description requirement would be applied by the Office for claims reciting a polynucleotide or polypeptide sequence that shares percent identity with another sequence (see "An Analysis of the New Written Description Training Materials - DNA Hybridization & Percent Identity").  We (and others) noted a curious aspect of the "new" way the Office would apply the requirement.  In the past, it was commonplace to recite claims having the following structure:

Claim 2:  An isolated nucleic acid that encodes a polypeptide with at least 85% amino acid sequence identity to SEQ ID NO: 2; wherein the polypeptide has activity X.

The limitation that the claimed sequences encompassed only functional species (polypeptides having activity X) was intended to exclude inoperative species.  Curiously, the new training materials instructed examiners to reject such claims unless the specification contained (or the art recognized) a structure-function relationship between the sequence and "activity X."  In explicating the Example, the training materials assert that where the specification discloses a single polypeptide sequence (or the putative sequence deduced from a cloned sequence), and without any disclosure regarding which amino acid residues could be altered whist retaining the recited activity, claims like the one above would fail to satisfy the written description requirement.

The following claim, on the other hand, would pass muster under the new application of the written description requirement:

Claim 1:  An isolated nucleic acid that encodes a polypeptide with at least 85% amino acid sequence identity to SEQ ID NO: 2.

The training materials justify this distinction because "[w]ith the aid of a computer, one of skill in the art could have identified all of the nucleic acids that encode a polypeptide with at least 85% sequence identity with SEQ ID NO: 2."

The more astute student of the present regime questioned whether there would be any other consequences for claims presented that did not recite any activity for the claimed polypeptide, and hence encompassed both operative and inoperative species.  And Dr. Elliott provided the answer:  while claim 1 satisfied the written description requirement, such a claim would fail to satisfy the enablement requirement.  When questioned after his formal presentation, Dr. Elliott further explained that the Office considers this equitable because the overwhelming majority of altered polypeptides would be inoperative, and it has long been the law that a claim encompassing an overwhelming number of inoperative species constitutes undue experimentation to be practiced by the public (once a patent's term has expired).

None of this, of course, is set out in the training materials (save the following "Practice Note" set forth at the end of the Example):

This, of course, is the tip-off, but it is more in the tradition of patent law being comprised of "traps for the unwary" than an administration trying to come to an equitable balance between claim scope and sufficiency of disclosure.  And while the Example cites some standard protein structure references, the Office is mistaken in the view expressed by Director Elliott about the consequences of amino acid sequence variation on protein activity.  Biologists have recognized for over forty years that most variation in most proteins is neutral, even when it has structural consequences like altering mobility on starch gels or changing a protein's pI.  Proteins, in short, are not so fragile, and that is consistent with the type of genetic variation needed to drive and sustain evolution.  More recently, the amount of genetic variation in an individual was found to be much higher than previously recognized, mostly due to insertions, deletions, and rearrangements in chromosomal DNA inherited from both parents (see "A Complete Diploid Human Genome Reveals Some Surprises").  The Office seems to believe that the fact that it is (typically) impossible to predict whether a particular change in a particular position will affect protein structure is enough to support a requirement that 35 U.S.C. § 112, first paragraph, requires explication of the effects of every change at every position.  This thinking is illustrated in a recent Board decision, In re Porro, which will be the subject of a future post but is enough to disturb the sleep patterns of any conscientious biotech patent practitioner (or applicant).

For now going forward, the most prudent course may be to disclose species orthologs or allelic variation (when known), to identify both invariant residues and residues that have been changed by nature.  Sequence alignments, or inclusion of "consensus" amino acid sequences in a Sequence Listing, should also be useful in addressing enablement rejections arising from presenting claims that satisfy the written description requirement as now applied.  And the Federal Circuit may yet resolve the issue, since one of the rejections appealed in Ex parte Kubin was for failure to satisfy the written description requirement in a claim reciting 80% amino acid sequence identity (and the Office's interpretation appears to violate the standard set forth by the Court in Enzo Biochem, Inc. v. Gen-Probe Inc.; see "Briefs for In re Kubin filed by Amgen and BIO").  Until this issue gets resolved, however, biotech patent applicants and their counsel can include the latest Patent Office "gotcha" game to the list of difficulties they can expect to encounter when trying to protect their inventions.

For additional information on this and other related topics, please see:

• "Docs at BIO: Panel Discusses Impact of USPTO Rules Changes and Patent Reform Legislation on Biotech Patenting," June 23, 2008
• "Docs at BIO: Representatives from JPO, EPO, SIPO, and USPTO Discuss Recent Developments in Japan, Europe, China, and the U.S.," June 22, 2008
• "Docs at BIO: Steve Burrill's State of the Biotechnology Industry Report 2008," June 19, 2008

Comments

[Inquiring Mind]

Thank you for the post. I read the Porro BPAI decision and am flabbergasted. Although I have not worked on this kind of claim for several years, this is the kind of rejection I was seeing, and I'm very sorry to see it succeeding with the Board (especially the biotech members of the Board). I hope that a resolution of Kubin will be helpful, but probably not; there has always been some tension between the obviousness standard and the written description standard for sequences, and for a long time practitioners were able to argue based on some quirky case law that biotech claims should get the better end of each argument. That is probably going to go away forever, but what will be left for sequence claims? Your tip to include consensus sequences is a good one, but given the way the PTO is applying the rules, I don't think that will help because the way they're articulating the rule, you're only going to get protection for the exact functional sequences you have disclosed in the application. If the rule of Porro is the final one, I don't see how anyone would be able to adequately protect an invention based on a sequence, since a design-around of a claim to an exact sequence is less than trivial. Hopefully the argument will be made that the standard being applied here to sequences is inconsistent with the standard as applied to other arts, and that Enzo and Lilly are inapt (as a special case and as one reciting only function, respectively) and common sense will prevail. Otherwise good luck to the industry in getting a legislative solution.