In re Chapman (Fed. Cir. 2010)

The Standard of Reviewing Obviousness Determinations from the USPTO

    By Kevin E. Noonan --

The Federal Circuit vacated and remanded a Board of Patent Appeals and Interferences determination that the claims in U.S. Application No. 09/719,045 were obvious on Thursday, in deciding In re Chapman.  The CAFC's decision addressed not only the proper obviousness standard that the Board must apply, but also revisited the proper standard of review the Federal Circuit applies regarding factual determinations by the Office according to the Supreme Court's instructions in Dickinson v. Zurko.

The Chapman application is a U.S. national phase of an International Application first filed in Great Britain.  The following claim is representative:

1.  A divalent antibody fragment comprising
    (a)  two antibody heavy chains and
    (b)  at least one polymer molecule effective for increasing the circulating half-life of said fragment in covalent linkage,
    (c)  each heavy chain being covalently linked to the other by at least one non-disulphide interchain bridge linking the sulphur atom of a cysteine residue in one chain to the sulphur atom of a cysteine residue in the other chain, said cysteine residues being located outside of the variable region domain of each chain, characterized in that the at least one non-disulphide interchain bridge contains the at least one covalently linked polymer molecule.

As the CAFC explained, the invention was directed towards making a derivative of an antibody comprising two heavy chains linked to one another using a polymer attached at each end to a sulfur atom comprising a cysteine residue in each of the different heavy chain molecules, and that the cysteine residues in each heavy chain came from outside the variable region domain (i.e., where the specificity of antibody binding is structurally determined).

Importantly for the Court's understanding of the underlying technology, the government explained that there was a need in the art for antibody fragments, since antibodies were frequently too large and had extended half-lives that could be deleterious to a patient.  Advantages of antibody fragments, according to the Court, were that they had better serum clearance rates and more rapid distribution throughout the body.  The Court noted that enzymatic digestion was known to produce antibody fragments effectively.  The Court also noted that one species (F(ab')2) produced by digestion (with pepsin) was "dumbbell shaped" and that it was divalent (having two antigen binding sites) as required by the claims.

With regard to the teachings of the specification supporting the Chapman claims, the Court said that Chapman's specification taught combining two separate Fab' fragments (having removed the light claims) "using an interchain bridge that contains at least one covalently linked polymer."  Thus, the linkage between two heavy chain molecules is mediated by the polymer rather than the disulfide bridges used in the native molecule.  One of the advantages of using a polymer rather than a disulfide bridge is that it increases the circulating half-life of the antibody fragment, according to the Court, to a half-life "that is intermediate between that of an individual fragment and a whole antibody."

The prior art asserted by the Office included U.S. Patent No. 6,025,158 (the Gonzales reference), which the Court said "describes linking antibody fragments to a polymer to increase an antibody's circulating half-life for therapeutic purposes."  One polymer known in the art, polyethylene glycol (PEG), when attached to the sulfur atom in the cysteine residue in the hinge region of the molecule "reduced clearance compared to the parental Fab' molecule"; the reference also describes dumbbell-shaped embodiments "made up of two antibody fragments linked by a polymer and the preparation of such conjugates," including Fab, Fab', Fab'-SH, F(ab')2, scFv, and Fv as choice for the antibody portion of the conjugate and PEG as the polymer linking them.  Included in the description of the '158 patent were methods for linking antibody fragments and the polymer without using a disulphilde bond in some embodiments (albeit using sulfur atoms in cysteine residues as the point of PEG attachment, specifically cysteine residues in the hinge region).

The Examiner applied the Gonzales reference for anticipation under 35 U.S.C. § 102(e), or in the alternative, combined Gonzales with U.S. Patent No. 5,436,154 to Barbanti (teaching antibody therapy) in rejecting the claims under 35 U.S.C. § 103.  The Board affirmed as to the obviousness rejection, reversing the anticipation rejection on the ground that "too much in the way of mental gymnastics would have been necessary for persons of ordinary skill to have 'at once envisage[d]' the claimed antibody structure among the different structures described in the Gonzalez patent.  . . .  [P]icking and choosing would have been necessary to have arrived at the antibody structure of claim 1."  The Federal Circuit noted several findings of fact made by the Board in affirming the Examiner's rejections.  These included:

• Antibody fragments disclosed in the Gonzales reference include monovalent Fab fragments, monovalent Fab' fragments having one or more cysteine residues in the constant region, or F(ab')2 fragments having a disulfide bond between cysteine residues in the hinge region.

• The Gonzales reference disclosed conjugates wherein the F(ab')2 fragment is conjugated to the polymer at a cysteine that would otherwise comprise the disulfide bond between the two fragments.  However, Gonzales describes the polymer as being linked to a cysteine in the light chain or the heavy chain, and that the cysteine in the other chain that would otherwise for a disulfide bond with the cysteine bonded to the polymer is changed to prevent a disulfide bond from forming.

• Gonzales in other embodiments describes a polymer that links two antibody fragments in a "dumbbell-shaped structure," using a polymer having multiple functional groups "permitting the direct attachment . . . of two or more antibody fragments to the polymer backbone."

The Board agreed with the examiner that the Gonzales reference had disclosed "linking two antibody fragments with a polymer to form a 'dumbbell-shaped' structure," and had expressly taught "a Fab molecule with a PEG linked to the hinge cysteine of the heavy chain."  "The only issue," according to the Board, was "whether persons of skill in the art would have had reason to join the [Fab'] fragments together using a polymer linked to the hinge cysteine residue."  The Board concluded that the preference disclosed in the Gonzales reference for linking the polymer to the hinge region would have "led the skilled artisan to utilize the claimed hinge cysteine," based in part on the advantages disclosed in the Gonzales reference regarding such linkages (i.e., producing "reduced clearance" from the bloodstream).  Especially important for the Board's decision was that, in its view, the Gonzales reference taught "a complete working example" of attaching a polymer to a hinge region cysteine residue.  The combination of this teaching, and the "dumbbell" structure disclosed in the Gonzales reference, "would have suggested to the ordinary skilled person that such Fab' fragments could be readily linked polymer to polymer using a bifunctional linker, as explicitly stated by Gonzalez when characterizing the dumbbell-shaped antibody structure."

In the Federal Circuit opinion, by Judge Dyk joined by Judges Gajarsa and Clevenger, the Court began with the standard of review:  no deference for legal conclusions, and substantial evidence for factual findings.  The opinion notes that both parties agree that whether the obviousness rejection was correctly maintained depends on whether the Office had correctly interpreted the teachings of the Gonzales reference.  In this regard, the Court refused to reverse the Board's determination as a matter of law, instead remanding to the Board "to correct certain errors in the Board's decision."  The Court rejected applicants' argument that the Gonzales reference teaches away from the claimed invention, based on the multiplicity of positions for polymer attachment, in view of the express preference for attachment at the hinge region.  The Court also noted that the reference was directed towards solving the same problem as the claimed invention -- increasing the half-life of the antibody in circulation.

Applicants' most persuasive argument was directed to three of the Board's factual findings, which both applicants and the government agreed were erroneous.  The difference in the parties position was that the Office argued that the errors were harmless, and the opinion hinged on the Court's assessment of the harm caused by these errors, applying the same "harmless error" rule that it applies to factual determinations from district courts, citing In re Watts, 354 F.3d 1362 (Fed. Cir. 2004) (citing In re McDaniel, 293 F.3d 1379 (Fed. Cir. 2002) and Gechter v. Davidson, 116 F.3d 1454 (Fed. Cir. 1997)), an approach the opinion asserted was affirmed by the Supreme Court in Shinseki v. Sanders, 129 S. Ct. 1696, 1706 (2009).

Specifically, the Board erred in finding that:

• Gonzales taught a "dumbbell-shaped" structure comprising two monovalent Fab' fragments lined via a polymer.  The government agreed that the Board "conflated" two correct statements on the record from the Examiner to arrive at an incorrect conclusion.  The Court found that this error was harmless, because the Board did not rely on "any such explicit disclosure" in the Gonzales reference

• "Gonzalez describes a divalent antibody in which the polymer is linked between light and heavy chains and only one cysteine residue is present."  The Examiner stated that "Gonzalez describes conjugates containing an 'F(ab')2 antibody fragment in which the polymer is attached between the disulphide bridge that would ordinarily link the light and heavy chains.'"  The government conceded at oral argument that applicants were correct, that the Gonzales reference in fact disclosed that:

[T]he conjugate contains a F(ab')2 antibody fragment attached to no more than about 2 polymer molecules, wherein every polymer molecule is attached to a cysteine residue in the light or heavy chain of the antibody fragment that would ordinarily form the disulfide bridge linking the light and heavy chains, wherein the disulfide bridge is avoided by substituting another amino acid, such a serine, for the corresponding cysteine residue in the opposite chain.

• Finally, applicants argued (and the government agreed) that Gonzales taught that there were up to six antibody fragments that could be conjugated to a polymer (F(ab), F(ab'), F(ab')-SH, F(ab')2, scFv, and Fv) and not just Fab, Fab' or F(ab')2 fragments.

The Court found that both of the latter references were not harmless because "they increased the likelihood that [applicants were] denied a patent on grounds of obviousness."  The Board's obviousness finding being based on "a misunderstanding" of the Gonzales reference "called into question" its legal determination of obviousness, according to the Federal Circuit.  Specifically, if the Board misinterpreted the Gonzales reference regarding whether it contained a teaching of using a polymer to link the light and heavy chains in an F(ab')2 fragment, then applicants' invention of using a polymer to link two F(ab')2 fragments together "may be less obvious" in the Court's opinion.  Similarly, with regard to the number of possible fragments that could be conjugated, "if the Board did not appreciate the full scope of antibody fragments disclosed in Gonzalez, we cannot be confident about its ultimate conclusion that the selection of one of them to form Chapman's molecule is obvious, as it appears that there are more possibilities from which to choose."  Thus, "[b]ecause we cannot say with confidence that the Board would have reached the same conclusion in the absence of these errors, we are persuaded they are indeed harmful."

Interestingly, despite the Court's clear understanding that the Board had made several prejudicial errors in making its obviousness determination, rather than reversing the Board outright the Court remanded to permit the Board to "revisit" its obviousness assessment, in view of a correct understanding of the Gonzales reference.  "The Board is in no way precluded from, and indeed may be correct in, finding the claims to be obvious, particularly in the light of Gonzalez's disclosure of joining two antibody fragments together with a polymer to make a dumbbell-shaped structure," wherein the Court thus exhibited a high level of deference to how the Board makes its factual findings.

In re Chapman (Fed. Cir. 2010)
Panel: Circuit Judges Gajarsa, Clevenger, and Dyk
Opinion by Circuit Judge Dyk

Comments

[EG]

Kevin,

Nice analysis of this case. I'm sure you've had instances like I've had where the examiner mischaracterizes what one or more references fairly teach, making the obviousness rejection problematic. I view Chapman as primarily important for holding the examiner's (as well as the BPAI's) "feet to the fire" that they fairly characterize what the reference(s) say, or expect any obviousness determination to be deemed suspect. After the KSR "common sense" nonsense view of obviousness, we need cases like Chapman to be able to challenge the factual correctness of the underlying characterizations of the references.

[6]

Kevin,

Nice analysis of this case. I'm sure you've had instances like I've had where the attorney mischaracterizes what one or more references fairly teach, making the obviousness rejections fairly simple. I view Chapman as primarily important for holding nobody's feet to any fire that they fairly characterize what the references say, or expect any obviousness determination to be deemed suspect. After the common sense KSR view of obviousness, we need cases like Chapman to remind us that you can spend a lot of money challenging something only to have wasted that money.

[Keep It Real]

Well played, 6. Of course, EG pitched you a softball right down the middle ...

[wanting more to chew on]

Awww, Kevin, I was hoping for your take on the ultimate question of obviousness. I've never dealt with antibodies so I didn't know how big of a difference bond type and shape are in the field. To me, it seems that the BPAI included the facts in the opinion it felt it needed to reject the claims. If those facts are wrong...

[EG]

6 and KIR,

I'm not saying that patent agents/patent attorneys don't also suffer from a similar issue. There are certainly instances where I've pushed the examiner to clarify what they believe the reference teaches and have realized that the examiner is on point so I don't push further. I also don't mind you disagreeing with my comment, but your respective responses don't advance the debate:

1. Even SCOTUS said in KSR International that an obviousness determination must be rationally supported by factual evidence (e.g., what the reference fairly teaches). If you don't believe me, note that SCOTUS cited with approval the passage in the Federal Circuit's In re Kahn case which basically says this. If the reference is mischaracterized for what it fairly teaches, you've got an evidentiary issue that, in view of Chapman, may undermine any obviousness ruling it's based on.

2. The fact is I've personally seen all too many instances where the examiner tries to make you guess at what the reference fairly teaches. One annoying tactic is to simply quote what’s in the claim (sometimes word for word) and then cite some passage in a reference that allegedly corresponds without ever identifying where the the specific claim elements are present in that passage. That doesn't even comply with 37 CFR 1.104(c)(2).

3. Another examiner favorite I've seen several times is to rely upon so much of the reference as supports the rejection, but then ignore other portions of the reference which teach away. I tell examiners who do this that the law is clear that you can't have it both ways.

4. One of the most hiliarious instances of mischaracterization of a reference I’ve experienced was an examiner who relied upon a drawing in a patent reference which appeared, at first blush, to be similar to a drawing in the patent application I was prosecuting. The problem was that each of the drawings represented entirely different and unrelated concepts, especially once you compared the text of the reference patent to the text in the patent application. Fortunately, the patent examiner got my “gentle hint” in my response that the reference drawing wasn’t as relevant as he thought it was, and dropped the rejection based on that reference drawing.

Again, I'm not dreaming up this mischaracterization of references issue that Chapman dealt with. In fact, other patent agents/patent attorneys have echoed what I've seen on other patent blogs.

[Kevin E. Noonan]

All:

One of the things I enjoy most about what we do is that it is always new - hard to get bored with a job where every time you open a case it's about something you (and almost everyone else) has never heard of before, at least for me.

But a consequence is that deciding how to claim a new invention properly, and deciding whether those claims are patentable, isn't easy. Thus the debate here and the problem in the case.

One of the worst reminders of the Dudas regime is the propensity - on both sides, examiners and applicants/attorneys - to demonize each other. Examiners are not (generally) lazy or incompetent; attorneys and applicants are not (generally) looking for invalid claims or to confuse the examiner or mischaracterize a reference. Most of the time, as in Chapman, there is a misunderstanding - a reference is honestly (but mistakenly) believed to disclose something that supports an argument of obviousness/non-obviousness (since anticipation tends to be (but is not always) a more yes or no situation).

We can all recite instances where the fact-finder isn't listening (remember the factual distinction Judge Rader did not credit in the Kubin case?), or an attorney doesn't/won't acknowledge a legitimate point and just amend the claims? I think the answer is to figure out a better way to make the point, or to suggest amendments that would avoid the reference. Expedite prosecution by determining what is patentable early, and then make sure that is on the record. Not a panacea, but "Examiner/applicant, you ignorant slut" isn't a very productive interaction.

Thanks for your comments.

[EG]

Kevin,

Sorry if I came off as arguing that examiners are somehow "ignorant." That wasn't my point. My point is that if a reference is mischaracterized (like in Chapman), you've got a fundamental problem with any obviousness determination based on that mischaracterization.

And you're correct this mischaracterization phenomena was much more prevalent during the Dudas regime. But it was a very real problem that I saw too many instances of to say it was isolated phenomena during the Dudas regime that we're still burdened with until the current Kappos reforms take effect (hopefully soon).

Once SCOTUS undermined the TSM test, accurate (and fair) characterization of the references becomes even more important if you want to argue a claimed feature/element isn't present in the cited reference. Put differently, a patent prosecutor can't simply accept what he/she considers to be a mischaracterization of the reference and then expect to argue or amend the claims successfully. That's one of the unfortunate legacies of the prior Dudas regime.

[Kevin E. Noonan]

Dear EG:

Didn't mean to scold; I understand the frustrations. And I agree that factual errors in interpreting what a reference teaches (which the PTO admitted had been made in the Chapman case) make it difficult to argue an obviousness rejection in the post-KSR world.

My point was that it should not be a hanging matter or a capital crime for someone (examiner or attorney) to make a mistake sometimes. If we start there, and if both sides spend less time snarling at one another, maybe we can more effectively get the process working again.

Thanks for the comment.

[EG]

Kevin,

Peace, brother patent attorney.

[6]

EG I was tempted to knock the next softball you pitched me out of the park as well but I'm going to refrain.

Let me simply say this instead: There is no debate. Mischaracterization goes on. Period.

The irony is, as I tried to point out to you through my little play on your post, that it happens on both sides so your btching is unwarrented. Every single statement you have made in this thread regarding the Topic can be turned around 180 degrees to the attorney side and still be true.