Brief mentions of IP stories floating around in the past few weeks that might be of interest.

-The Supreme Court continues to offer the last word on many patent issues, already issuing two patent decisions this term, with AMP v. Myriad soon to be added to that list, and showing interest in several other patent issues. [PatentlyO]

-The surge in application filings in March associated with changes brought about by the AIA has caused the USPTO backlog to increase to levels last seen six months ago. [USPTO Data Visualization Center]

-Until September 30, the USPTO has brought its After Final Consideration Pilot back to life. Will it bring many applications back to life after final? [USPTO AF Consideration Pilot]

-This. This is what an indefinite claim looks like. [In re Hartman]

-EFF tries to kill patents on 3-D printing, but will 3-D printing try to kill patents? [TechDirt]

-Speaking of which, if a 3-D printer prints another 3-D printer, is this self-replication not subject to patent exhaustion? [SCOTUS Bowman v. Monsanto opinion]

Case No. 2011-1547 (Newman, O’Malley, Wallach)

Federal common law is alive and well in patent jurisprudence. The courts have created several patent doctrines without any statutory basis, and those that do have statutory backing often require thousands of words in judicial opinions to understand and revise (“any new and useful process” comes to mind). This second post on managing patent families (first post here) deals with a doctrine from the former category, the appropriately-named nonstatutory double patenting doctrine. The issue is a common one amongst related applications, as seen in the recent In re Hubbell.

The history in Hubbell was Greek-tragedy-level complex. Put simply, two application families – one to a genus and the other to a species – had overlapping inventorship but different ownership. The first family, owned by CalTech, included the application on appeal, 10/650,509, and had priority to a provisional application filed in 1997. The ‘509 application claimed a genus of “[a] bidomain protein or peptide comprising a transglutaminase substrate domain and a polypeptide growth factor.” While the ‘509 application family was in prosecution, two of its inventors left CalTech for ETHZ, where they filed the other family in 1998, ultimately yielding US Patent 7,601,685 in 2009, which claimed a species of the ‘509 application’s transglutaminase substrate domain, “crosslinking Factor XIIIa substrate domain.” Although the ‘685 patent was not valid prior art against the ‘509 application due to its later priority date, the USPTO rejected the ‘509 application genus claims under nonstatutory obviousness-type double patenting over the ‘685 patent species claims because of the overlapping inventorship. An appeal to the Federal Circuit ensued. . .

The ’509 applicants made three arguments on appeal from the Board decision: 1) obviousness-type double patenting should not apply between applications without any co-ownership; 2) the USPTO should accept a terminal disclaimer to overcome the rejection; or 3) double patenting here should require a two-way analysis. The panel majority accepted none of these. In refuting these arguments, the majority was particularly concerned about multiplicity of infringement:

There are two justifications for obviousness-type double patenting. The first is “to prevent unjustified timewise extension of the right to exclude granted by a patent no matter how the extension is brought about.” Van Ornum, 686 F.2d at 943-44 (quotation and citation omitted). The second rationale is to prevent multiple infringement suits by different assignees asserting essentially the same patented invention. Fallaux, 564 F.3d at 1319 (recognizing that “harassment by multiple assignees” provides “a second justification for obviousness-type double patenting”)

The Van Ornum and Fallaux courts had both been confronted with applications that were, at some point, co-owned. The original owner in those cases had itself created the multiple-ownership, and thus multiple infringement, issue through assignment. Here, CalTech had no control over the ‘685 patent at any time and could not be said to have created or avoided the multiple-ownership problem. Nonetheless, because the split ownership of obvious claims could result in double jeopardy for an infringer, the majority thought obviousness-type double patenting should be extended to the facts here:

It is undisputed that the ’685 patent is assigned to ETHZ and Universitat Zurich, whereas the ’509 application is assigned to CalTech. While Hubbell is correct that CalTech was not responsible for the bifurcated assignment, the fact remains that, if Hubbell’s rejected application claims were to issue, the potential for harassment by multiple assignees would exist because an infringer of claim 1 of the ’685 patent would also infringe at least rejected claim 18. As such, there is a risk that a potential infringer could be subject to suit from both CalTech and ETHZ and Universitat Zurich under their respective patents. The fact that Hubbell controlled the assignment by moving from one institution to another does not alter this analysis.

Although Hubbell attempts to distinguish Van Ornum and Fallaux on grounds that the applications involved in those cases were once commonly owned, this distinction does not alleviate the underlying harassment concerns here. As the Director points out, the “possibility of harassment by multiple assignees exists when issued patents lack common ownership. Thus, whether the applications that led to the issued patents were ever commonly owned is irrelevant.” Appellee’s Br. 16. And, as we recognized in Fallaux, the harassment justification is “particularly pertinent” in cases where, as here, the application and the conflicting patent are not commonly owned.

On the ability to file a terminal disclaimer, the majority again pointed to the infringement-harassment-by-multiple-parties problem. Terminal disclaimers required that the conflicting applications be commonly-assigned to avoid this exact circumstance. Thus, a terminal disclaimer was not eligible and would not address the plural infringement concerns here. The applicants on appeal noted that a party to a joint research agreement could file a terminal disclaimer that promised not to separately enforce the disclaiming application, but the court found this solution was provided for by statute through the CREATE ACT, whereas non-related assignees were not:

Although Hubbell argues that this court should ignore the joint research agreement requirement and nonetheless permit the filing of a terminal disclaimer, there is no statutory basis for doing so, and we decline Hubbell’s invitation to rewrite the statutory text by creating a new equitable right to resort to terminal disclaimers. We conclude that, because Hubbell does not qualify for a terminal disclaimer under the statute, he is not entitled to file one as an equitable measure.

On the two-way test, which would require the conflicting claims be mutually obvious over each other before an obviousness-type double patenting rejection could be lodged, the panel majority found that this test was reserved for faultless applicants who had not contributed to inverted issuance of previously- and later-filed applications. Thus, only a single-way test was required here. In so holding, the majority noted that Hubbell shared some of the blame for the ‘509 application still being in prosecution while the ‘685 patent issued. Hubbell had admitted that some of the responsibility for the prosecution delay was not the USPTO’s and that the genus claims on appeal could have been presented (and likely allowed) in earlier applications in the ‘509 family. This was enough to preclude a two-way test.

Judge Newman dissented. Judge Newman would not have extended Van Ornum and Fallaux to the facts here, because double patenting had always required either identical inventorship or co-ownership at some point during pendency of the conflicting applications.

Nonstatutory obviousness-type double patenting rejections affirmed.

Lessons: I think the best lesson from Hubbell is for portfolio managers to keep an eye on moving inventors and their patenting activities; don’t rely so much on employee agreements to protect against conflicting application families. When inventors move on to another R&D position that is likely to generate patent applications related to their work with you, it is critical to timely present all claims to subject matter they invented to the USPTO and expeditiously prosecute them. As seen in Hubbell, a departed inventor’s improvement applications at their new shop (especially if prioritized) can come back and kill your claims without remedy if you do anything that might be considered “delay.” As long as you present and expeditiously prosecute the claims, you’ll at least be afforded the two-way test remedy. It also might be worthwhile (and evil) to carefully review past patent work of incoming inventors from a competitor and prioritize their improvement applications; this may cause an unvigilant competitor to lose patent rights or at least finalize all desired claims under a rush basis under Hubbell.

As for my skepticism on employee IP rights agreements, in Hubbell, for example, I think it would have been difficult to draft an agreement that would have gotten the ‘685 patent back under the ownership of CalTech. And getting the ownership back to CalTech was the only way to submit the terminal disclaimer. The ‘685 claims were pretty inventive over the ‘509 claims, such that the two departing inventors likely did quite a bit of non-obvious work after leaving CalTech. I think it would be difficult to contractually force a departed inventor who comes up with and claims a critical species or non-obvious improvement well after employment termination to assign all future application with such claims back to a previous employer. An employment agreement requiring assignment of any application that is the basis of a DP rejection back to the previous employer is doing just that. These start to resemble unenforceable perpetual non-compete clauses.

Comment: The majority opinion is unconvincing to me. So much of the majority’s rationale and analysis was based on their belief that multiple infringement suits are unacceptable and thus the DP rejection must be maintained. But we tolerate multiple infringement suits between base and improvement patents all the time; the majority offers no convincing reason why multiple infringement is anathema only if an inventor is shared. If the ‘509 application had completely different inventorship (or merely presented the genus claims early enough), there would have been no DP rejection and someone practicing the ‘685 claims would have infringed both patents held by different parties. Why do our multiple infringement alarms go off when a single inventor is shared and she does not timely prosecute her claims to the hilt? There may be a reason for this, but the majority doesn’t offer it.

Moreover, why not allow a joint-research-agreement-style terminal disclaimer here, promising no separate enforcement of the differently-owned applications? The majority opinion relies on there being no statutory basis for such a terminal disclaimer filing when a joint research agreement is absent, but there’s also no statutory provision for the joint-research-agreement-style terminal disclaimer (it’s created by regulation). Moreover, for decades, there was zero statutory basis for terminal disclaimers to require co-assignment, but the court allowed it through opinion and the USPTO enforced it through regulation. There’s certainly no statutory prohibition, so why require explicit statutory acceptance now for an equitable remedy that would have addressed the majority’s biggest problem – disparate enforcement?

On a side note, the opinions presume that full priority support is present in both families, such that the ‘685 patent was not statutory prior art to the ‘509 claim. But the opinion includes this passage as evidence of delay in presenting the claims at issue until 2003:

“JUDGE FREDMAN: “This isn’t a situation where the PTO prevented you from presenting these claims way back in 1998 or whatever it would’ve been when you had the first case.” MS. MONHEIT: “The earliest point that we could have presented [the claims on appeal] is when we filed that CIP, which is in 2001.”

I’m not going to go digging through the disclosures and history of these families to determine support, but doesn’t this exchange suggest the ‘509 claims only had priority to 2001, which was after the 1998 and 2000 priorities of the ‘685 patent? That would make the ‘685 patent statutory prior art. Maybe the ‘685 patent claims actually had later priority, as they were in a CIP application?

Lastly, this is outside the facts of Hubbell, but still an interesting hypothetical: what if the later ‘685 claims were obvious over, or simply a previously-identified species within, the ‘509 claims? Would the USPTO still issue a DP rejection against the ‘509 claims, and, if so, this seems really unfair because the ‘685 owner could simply obstruct the ‘509 application with junk claims, right? Answer: No. The ‘685 claims would be rejected, at least under 102(e), over the ‘509 application, and, at best, provisional DP rejections could be made in the ‘685 and ‘509 applications. That is, no ‘685 patent would have issued with such claims. Once the provisional DP rejection was the final rejection remaining in the ‘509 application, it would have been withdrawn under MPEP § 804(I)(B)(2). So no hold-up or DP kill unless the ‘685 claims are truly patentable over the ‘509 application, which they were in Hubbell

Case Nos. 2010-1360 (Rader, Newman, Moore)

Happy patent families are all alike; every unhappy patent family is unhappy in its own way. (I’m sure this is how Tolstoy really meant it.) Properly managing a portfolio of related applications is difficult. The incremental nature of invention, vagrant inventors, and constraints in IP harvesting often make for tough decisions in dividing up subject matter between related applications, determining proper priority chains, and providing correct inventorship for each application in a family. Complex and occasionally inconsistent Federal Circuit caselaw does not make these jobs easier. This series looks at two cases dealing with problems in patent families: this post will look at last fall’s Santarus v. Par Pharma for issues in priority claiming; the next post will deal with the more recent In re Hubbell for double patenting troubles.

Santarus owned a family of patents covering the drug Zegrid, including US Pats. 6,489,346; 7,399,772; 6,780,882; 6,699,885; and 6,645,988, each of which sprang from an ancestor patent 5,840,737 as continuations or continuations-in-part. Par Pharma filed an ANDA, creating a Paragraph IV challenge to the validity and enforceability of the patents. The challenge was successful; the district court found several claims of the ‘346, ‘885, and ‘998 patent obvious over the ancestor ‘737 patent itself, to which the invalidated patents claimed priority. Santarus appealed these findings. . .

Could the ’737 patent be prior art against the only partially-supported child patents? Unfortunately, the majority on appeal found that Santarus had waived any argument that the ‘737 provided priority for the invalidated claims (except for claim 1 of the ‘772 patent, discussed in the second section below). Thus, we get no discussion from the majority as to whether a parent application claimed to under § 120 by a continuation-in-part can be prior art against the claiming application. In re Chu is probably the best stand-in for the position that, yes, claims in a C-I-P can be invalidated by the priority application, if the priority application does not completely support the claims.

Judge Newman, apparently finding the issue not waived, does provide contrary commentary about C-I-P priority in a dissent. It was the dissent’s view that a claim of priority to any ancestor patent eliminates the ancestor patent from the prior art, regardless of whether the ancestor patent supports the claim. That is, as long as a child application has overlapping inventorship with, is filed during the pendency of, and includes a reference to a parent, the subject matter of the parent is off limits as prior art. As Judge Newman put it:

The panel majority forgets that “matter disclosed in the parent application is entitled to the benefit of the filing date of the parent application.” Waldemar Link, GmbH & Co. v. Osteonics Corp., 32 F.3d 556, 558 (Fed. Cir. 1994); see Litton Sys., Inc. v. Whirlpool Corp., 728 F.2d 1423, 1438 (Fed. Cir. 1984) (“The earlier filing date of the parent application pertains to material in the C-I-P application also disclosed in the prior application. 35 U.S.C. §120.”). Instead, the panel majority relies upon the common subject matter from the ’737 patent disclosure to invalidate the ’885 claims supported by that subject matter. This is incorrect, for the common subject matter in the ’885 patent is entitled to the ’737 filing date. That entitlement is not lost by issuance of the ’737 patent. The common subject matter, properly carried forward in copending continuing patents, cannot be prior art against itself, as the majority holds.

The majority, treating the ‘737 patent as proper prior art against claims in the continuation-in-part applications, further agreed that the disclosure in the ‘737 application rendered these claims obvious (The substantive obviousness discussion might be discussed in a future post). Invalidity affirmed as to the claims for which the ‘737 patent was prior art.

Lessons: Continuation-in-part practice might look like a good idea for improvement or related disclosures, but it offers mostly illusory benefits. As seen in Santarus, a C-I-P priority claim will not eliminate parent applications as prior art if they do not fully support the C-I-P claims (and if they did, you could simply claim straight continuation priority). And C-I-P practice has real drawbacks, like needless loss of patent term and false assurances of validity. So the take-away is that priority is best, and most safely, claimed only between true continuation and divisional applications having the same disclosure or at least fully supporting the claims of the child application.

Nonetheless, I think a C-I-P has some usefulness when it’s not clear whether a parent application provides full § 112 support for a child claim. This can happen in a slowly-evolving project that has incremental improvements, as well an in large system inventions having several separately-patentable parts, where different subject matter can get spread across multiple applications. The uncertainties of BRI, litigation claim construction, and written description jurisprudence can make determining claim support at filing rather fuzzy, and the determination can easily vary from claim-to-claim. A C-I-P priority claim at least preserves the option of relying on priority for a subset of supported claims, which may become much more apparent during prosecution. If following this approach of making a C-I-P claim in light of uncertain support, make sure the Examiner is aware of potentially differing claim priorities and properly policing it – both in according the earlier filing date for supported claims and applying parent applications and intervening prior art to unsupported claims – to avoid the PowerOasis problem of specious validity. Also confirm that you’re ok losing patent term on any unsupported claims that are nonetheless found patentable over the parent and intervening art (and, if not, consider dividing supported and unsupported claims into the existing and a new application and appropriately deleting priority claims to maximize term).

Coming at the issue from the other angle, recognize that in some situations you don’t even need priority, continuation or C-I-P, between related applications (“related” meaning overlapping inventorship and co-owned or jointly-researched ownership). Due to publication delay, pre-AIA, applicants could get up to 30-months between filing dates of related applications with no priority claim before the original US filing became prior art to the later filing. Post-AIA, this window shrank to 18 months. (These windows can be extended through non-publication requests for US-only, non-prioritized filings.) If you are the only innovator in the narrow field of the family and not otherwise disclosing/commercializing before publication, such that it is only parent applications and not intervening disclosures that are prior art concerns, priority claims within these windows mostly just kill patent term because the parent will be ineligible or removable as prior art through other means. Moreover, if you find out during prosecution that the priority claim was wrongly omitted, you can add it late for a hefty fee as long as you have co-pendency and inventorship overlap. I’ve seen this route actually be less expensive than preparing a 131 declaration for antedating an anticipating earlier application/disclosure in complex families pre-AIA. In sum, even if you are in one of the above-mentioned scenarios where a continuation or C-I-P claim is useful, think very hard before making any such claim before the parent has published.

Comment: Judge Newman, to me, seems to be quite out on a limb arguing that the non-supporting parent application, to which priority is otherwise properly claimed, cannot be prior art for all that it teaches. She took a similar position in Tronzo v. Biomet, stating in dissent that the unsupporting parent application eliminated other intervening prior art having the same disclosure as the parent application to which priority was claimed. These seem at odds with Chu, for which Judge Newman was a panelist. The cases cited by the dissent (Waldemar Link and Litton Sys.) are also not on-point for the dissent’s position, and older cases like Stryker or Stempel discussed on other comment threads are rather limited to the swearing back milieu where they discuss eliminating less than fully-supporting references from the prior art. What’s more, in her concluding paragraphs Judge Newman seems to suggest that the USPTO always ensures that written description for priority exists, which is contrary to the reasoning in PowerOasis. I don’t think Judge Newman’s take on this issue is necessarily bad policy-wise, considering patent term is now from claimed priority filing date and double patenting will snuff out attempts at de facto term extension, but I don’t think it lines up with precedent.

* * *

Since we’re on the issue of priority, it’s worthwhile to take a look at a related in Santarus: written description support for a negative claim limitation. The youngest patent in the family, ‘772, included a claim 1 with a negative limitation of “wherein the composition contains no sucralfate.” The ‘737 parent application mentioned sucralfate as a known ingredient for treatment of gastric disorders, but in the context of causing problems (well, death) in the background. The district court thought that this was insufficient to provide written description to the negative limitation. The Federal Circuit disagreed and found that this mention, even though not in connection with any example embodiment, was sufficient to provide written description for the negative limitation and thus priority to the ‘737 parent application. As the majority put it:

Santarus argues that it is not necessary to include in the specification evidence of “contraindication” of sucralfate. . .

We agree. The claim limitation specifying that sucralfate is not administered in conjunction with the Phillips formulation restricted the claims to this preferred use of the Phillips formulations. This exclusion narrowed the claims, as the patentee is entitled to do. The Manual of Patent Examining Procedure explains that claims may state the exclusion of alternatives. See MPEP § 2173.05(i) (“If alternative elements are positively recited in the specification, they may be explicitly excluded in the claims.”).

. . . Negative claim limitations are adequately supported when the specification describes a reason to exclude the relevant limitation. Such written description support need not rise to the level of disclaimer. In fact, it is possible for the patentee to support both the inclusion and exclusion of the same material. The claim limitation that the Phillips formulations contain no sucralfate is adequately supported by statements in the specification expressly listing the disadvantages of using sucralfate.

Judge Newman in concurrence did not like the majority’s phrasing. Particularly, she pointed out that there is no requirement the specification disclose any reason to exclude the subject matter of the negative limitation. All that is required is that the specification identify the subject matter:

The applicant’s obligation is to describe and claim the invention in accordance with 35 U.S.C. §112. Thereafter, patent examination may lead to amendments to the claims. The MPEP §2173.05(i) advises that: “If alternative elements are positively recited in the specification, they may be explicitly excluded in the claims.” The MPEP does not require that the reason for such exclusion must be stated in the specification. The panel majority creates a new and far-reaching ground of invalidity, a ground that received no deliberation and advice from the concerned communities.

Because the ‘737 disclosure supported the claim of the child ‘772 patent, including the negative limitation, the ‘737 patent was not prior art against the ‘772 patent, and the claim was not invalid for lacking written description. The invalidation of this claim was reversed and remanded.

Lessons: A negative claim limitation can likely be supported by disclosure of the subject matter of the same in the specification merely as an alternative or by disclosure of the subject matter, even in the background, as something to be avoided. I do not read the Santarus majority as undercutting the rule of MPEP § 2173 dealing with disclosure of alternatives; rather, I think the majority expands the bases for written description support to include a disclosures unrelated to any example embodiment that nonetheless denigrate the subject matter sought to be the object of a negative limitation. Thus, all the more reason to disclose as many alternative species as possible in the specification, and explicitly state that any combination of disclosed elements are useable in example embodiments. The expansion by the majority also works well with problem-and-solution statements in the specification, as this might provide support for a negative limitation disclaiming identified particulars of the problem.

A short caution, however, on negative limitations in general. Distinguishing prior art with a negative limitation alone can be less than productive; it can feel like (and sometimes is) a cheap dodge to which an Examiner will respond to by throwing in a combinable reference also lacking the feature and/or an obvious reason for the omission. Negative limitations also receive somewhat higher scrutiny abroad. The EPO and JPO have several complex guidelines for treating “disclaimer,” and I’ve found none of them as kind as the US rule.

Comment: Again, I think the majority is correct here, and I have some trouble understanding Judge Newman’s response and alarm. I don’t see Santarus as dealing with the situation of disclosure of alternatives in the detailed description, or the majority as attacking the same, so I’m not sure why Judge Newman felt like this basis for negative claiming required defending in a separate opinion.

Case No. 2012-1120 (Dyk, Plager, Reyna)

Gotcha! Prosecution history disclaimer can feel like an unfair trap for practitioners; statements made years earlier in prosecution are exhumed and reanimated in litigation to show that the patentee surrendered patent rights before the USPTO. It thus behooves prosecutors to watch for fact patterns that courts associate with disclaimer and avoid repeating those patterns in their practices. The Federal Circuit, however, does not always speak consistently as to what type of arguments made in prosecution can later be used to narrow claim constructions in litigation. Witness Biogen Idec v. GlaxoSmithKline. . .

Biogen Idec owned US Patent 7,682,612 directed to use of anti-CD20 antibodies as a cancer treatment. CD20 is an antigen that is well-expressed on white blood cells called B cells. Anti-CD20 antibodies bind to the CD20 antigens, starting a cascade that kills the B cell. This is a useful therapy for disorders that cause a patient to have too many B cells, like leukemia. Several different types of antibodies have anti-CD20 activity, including human, humanized, and chimeric antibodies, and the ‘612 patent claimed them all, generically reciting “administering an anti-CD20 antibody . . . in an amount effective to treat the . . . leukemia.” A dependent claim narrowed the anti-CD20 antibody to a “chimeric, human, or humanized antibody.” Biogen’s actual product, Rituxan, used only a chimeric antibody that had affinity for only a particular type (epitope) of CD20 antigen.

Later, GSK developed and began marketing Arzerra, also an anti-CD20 antibody to treat the same disease classes as Rituxan. But Arzerra used a human antibody that had affinity for a different epitope of the CD20 antigen. Biogen asserted the ‘612 patent against GSK, based on claim language that was generic to antibody type. The district court, however, construed the claims of the ‘612 patent narrowly, based on this supposed disclaimer in the file wrapper:

During prosecution, the Examiner rejected the claims, which she interpreted as broadly covering all antibody types, as non-enabled, because the specification discussed only a few specific Rituxan-related antibodies. Biogen successfully traversed the rejection in the following manner:

Examiner:
Claims 1 and 12 are broadly drawn to ‘. . . an anti-CD20 antibody or fragment thereof’. This is broadly interpreted for examination purposes to be any and all anti-CD20 antibodies, no matter the specificity or affinity for the specific epitope on the circulating tumor cells. While the specification is enabling for the application of RITUXAN®, RITUXIMAB® and 2B8-MX-DTPA in the treatment of hematologic malignancies, the specification is not enabling in the application of all other anti-CD20 antibodies, which may have different structural and functional properties.

Applicants:
Applicants respectfully submit that even though antibodies directed to the same antigen might have different affinities and functional characteristics, one of skill in the art could readily identify an antibody that binds to CD20 with similar affinity and specificity as does RITUXAN® using techniques that are well known in the art. . . . With that knowledge in hand, the skilled artisan could readily produce anti-CD20 antibodies using similar techniques, and screen such antibodies for those having an affinity and functional activity similar to Rituxan®.

On appeal, a majority of the Federal Circuit panel agreed that this was prosecution disclaimer of non-Rituxan-like antibodies. From the outset, the majority took a rather favorable view of prosecution history disclaimer. They reasoned that the public is entitled to rely on the entire history of a patent in determining its scope of protection and that statements made in prosecution reflect claim terms’ proper definition. The majority put it like this, with heavy citing (indeed, the whole opinion cites a large amount of caselaw):

Prosecution history disclaimer plays an important role in the patent system. It “promotes the public notice function of the intrinsic evidence and protects the public’s reliance on definitive statements made during prosecution.” Id. Such statements can take the form of either amendment or argument. See Elkay Mfg. Co. v. Ebco Mfg. Co., 192 F.3d 973, 979 (Fed. Cir. 1999); see also Computer Docking Station Corp. v. Dell, Inc., 519 F.3d 1366, 1374 (Fed. Cir. 2008) (“Statements made during prosecution may also affect the scope of the claims.”). For this reason, the entirety of a patent’s file history captures the public record of the patentee’s representations concerning the scope and meaning of the claims. Seachange Int’l, Inc. v. C-COR Inc., 413 F.3d 1361, 1372 (Fed. Cir. 2005) (quoting Hockerson-Halberstadt, Inc. v. Avia Grp. Int’l, Inc., 222 F.3d 951, 957 (Fed. Cir. 2000)); see also Elkay, 192 F.3d at 979 (“[I]t is the totality of the prosecution history that must be assessed, not the individual segments of the presentation made to the [PTO] by the applicant . . . .”). Competitors are entitled to rely on those representations when determining a course of lawful conduct, such as launching a new product or designing-around a patented invention. Id. Beyond the notice function and reliance based aspects of a patent’s prosecution history, it “provides evidence of how the [PTO] and the inventor understood the patent.” Edwards Lifesciences LLC v. Cook Inc., 582 F.3d 1322, 1327 (Fed. Cir. 2009) (quoting Phillips, 415 F.3d at 1317) (alteration in original).

Here, the majority found there had been clear disclaimer. Specifically and importantly, the Examiner had understood the claims to be enabled only for antibodies that bind in a specific way to CD20 and rejected the claims for reciting more than that. Biogen’s above-quoted response adopted a narrow construction of the claims as limited to the Rituxan-like antibodies, without discussing other antibodies that had different binding properties, in order to overcome the rejection. In the majority’s thinking, if other antibodies with different affinities from Rituxan were to be included, the response would have talked about them instead of partially repeating the narrow claim aspects the Examiner identified as enabled:

rather than challenging the examiner’s understanding of the crucial terms, the applicants argued that the specification was enabling for anti-CD20 antibodies with similar affinity and specificity as Rituxan®. Indeed, the applicants conceded that other “antibodies directed to the same antigen [i.e., CD20] might have different affinities and functional characteristics,” and limited their claims to antibodies similar to Rituxan® nonetheless. Joint App’x 324. While the applicants may not have repeated the examiner’s language verbatim et literatim, it is clear that they were limiting their invention to what the examiner believed they enabled: antibodies that have a similar specificity and affinity for the specific epitope to which Rituxan® binds.

. . . The examiner began by characterizing antibodies by their specificity and affinity for a specific epitope, and the applicants adopted that characterization when they limited their claims to antibodies similar to Rituxan®. While disavowing statements must be “so clear as to show reasonable clarity and deliberateness,” Omega, 334 F.3d at 1325, this requirement does not require the applicant to parrot back language used by the examiner when clearly and deliberately responding to a particular grounds for rejection. If an applicant chooses, she can challenge an examiner’s characterization in order to avoid any chance for disclaimer, but the applicants in this case did not directly challenge the examiner’s characterization. See TorPharm Inc. v. Ranbaxy Pharm., Inc., 336 F.3d 1322, 1330 (Fed. Cir. 2003) (“Whether the patentee chooses to dispute the examiner’s view of matters is relevant to claim interpretation, for there a court may need to ascertain exactly what subject matter was actually examined and allowed by the PTO.”). Rather, they simply discussed specificity and affinity with regard to the disclosure of the ’612 patent, which was narrowly limited to Rituxan®, rituximab, and 2B8-MX-DTPA.

Having found that the patentee clearly disclaimed all other types of antibodies aside from Rituxan-type antibodies that acted on a specific CD20 epitope, the majority also dismissed arguments of claim differentiation and specification broadening. The majority noted that claim differentiation was merely a presumption that can be overcome by disclaimer and, anyway, the dependent claims to humanized or human antibodies could apparently (?) still be met by antibodies similar to Rituxan (the majority gave no reasoning as to how). Where the specification incorporated another patent that broadly defined “anti-CD20 antibody,” the majority reasoned that definitions from incorporated documents using terms like “as used herein” did not carry over into incorporating patents, and, even if they did, the prosecution disclaimer overcame those redefinitions.

Judge Plager dissented, finding no disclaimer. The dissent pointed to ambiguity in the above-quoted response and saw no clear statement that all antibodies except Rituxan and its equivalents were outside the scope of the claims:

The examiner was addressing an enablement issue; the applicants’ response was at worst a non-response to the examiner’s concern, and at best a statement that antibodies other than RITUXUN® and RITUXUN®-like antibodies that had similar affinity and specificity are included in the claims. “Similar” does not mean “same.” Applicants’ usage of “similar” is inconsistent with an acceptance by the applicants of the narrow confines proposed by the examiner’s reference to the “specificity or affinity for the specific epitope.”

. . . it is possible to argue, as GSK has done, that the “anti-CD20 antibodies” that may be used in the claimed methods are limited to only RITUXAN®-like antibodies. But it is at least equally possible, and more correct, to read the applicants’ prosecution arguments as contemplating the use of antibodies other than RITUXAN®-like antibodies. And when a prosecution argument is subject to more than one reasonable interpretation, it cannot rise to the level of a clear and unmistakable disclaimer. 01 Communique Lab. Inc. v. LogMeIn, Inc., 687 F.3d 1292, 1297 (Fed. Cir. 2012). We have long followed the rule that even a poorly-phrased prosecution argument does not a disclaimer make.

The dissent further noted that the Examiner introduced the epitope-dependent binding property; Applicants never repeated or otherwise addressed epitope-type in their response. When an Examiner unilaterally introduces a narrow or wrong claim interpretation, there typically is no disclaimer. Moreover, other portions of the prosecution history and specification made it clear that any anti-CD20 antibodies were useable in the claimed methods, and the broad definition of anti-CD antibody from the incorporated document was specifically relied on in the supposedly disclaiming response to the enablement rejection. Judge Plager would have remanded with the broader, plain language definition of “anti-CD20 antibody.”

Affirmed.

Lessons: Biogen is very aggressive in finding prosecution history disclaimer. Although I’ve previously suggested that merely parroting back an Examiner’s rejection and then distinguishing or making other arguments will not give rise to disclaimer, the majority in Biogen seems to want more out of applicants. Thus, consider identifying in the response any relevant and overly-narrow formulation by an Examiner as incorrect, at least to the degree permissible while still overcoming a rejection. Keep in mind that disclaimer can lurk within a single statement taken out of context from the remainder of the application, and in connection with any traverse, not simply one distinguishing prior art. The patent prosecutor’s night is dark and full of terrors indeed; consider pushing back against any Examiner statement being traversed that could be seen to represent a narrow understanding of the claims.

The above lessons being said, I admit that calling out and correcting every possible Examiner misstatement bearing on claim scope in a response is often not feasible and will not create a very good rapport with the Examiner or make responses more concise or comprehensible. So a more conservative lesson might be to try to avoid prosecution history altogether. Try to present at least one unique claim or claimset that is fairly narrowly drawn to the disclosure, with its own terminology, from the outset and will thus likely be allowed without rejection. This not only avoids prosecution history disclaimer in connection with this claim, but also estoppel for DoE purposes (and it often gives an early indication of how aggressive an Examiner is / is going to be in prosecution).

Comment: I think the dissent has the much better position here. There may have been problems with giving Biogen the broad claim construction under the claim plain meaning (including enablement still), but using prosecution history disclaimer to avoid that problematic scope feels very square-peg-round-hole. The majority awkwardly stretches to find disclaimer at several points: relying on epitope-specific characteristics introduced by only the Examiner; characterizing a rather bland traverse as a sub silentio adoption of the reverse of the Examiner’s characterizing; not really explaining how claim differentiation could still be honored here (human antibodies with only Rituxan-like epitope preferences?); and suggesting that a relied-upon incorporation by reference of another patent will not carry forward broader definitions of claim terms from that patent.

The Biogen majority arguably advances a rather new presumption of “prosecution history disclaimer unless rebutted on the record by the prosecutor.” This does not seem consistent with caselaw requiring disclaimer to arise only from an initial explicit and clear disavowal of claim scope from the prosecutor, and disclaimer is an area where prosecutors really need consistent guidance to respond appropriately and without disclaimer during prosecution in order to avoid malpractice. Worse, if this presumption of disclaimer is the new consistent rule, I think we will see prosecution responses grow in bulk from “nothing herein is a disclaimer” boilerplate and hyper-vigilant rejoinders to every potentially erroneous treatment of claim scope in an office action. I sometimes worry that the patent jurisprudential development process does not adequately account for the realities of practice before the USPTO, and Biogen does not help on this front.

Was the Plager dissent originally the majority opinion? The dissent reads at points like a remand, especially in conclusion and footnote 3. The majority opinion footnotes 5 and 6 addressed citations in Judge Plager’s opinion in somewhat non-sequitur fashion and feel like repurposed portions of a dissent. Maybe either Judge Dyk or Reyna flipped during opinion circulation?

Case No. 2012-1609 (Rader, Lourie, O’Malley)

While mere attorney argument cannot take the place of evidence before the USPTO, evidence does not always have a mandatory place in traversing factual matters before the Office. The Federal Circuit in In re Morsa finds attorney argument alone sufficient to traverse a rejection on the rather factual question of enablement, requiring substantive consideration and response from the Office. The case thus provides at least one set of facts where an Applicant should push back against a summary dismissal of a traverse as “mere attorney argument.” Morsa is also a nice companion to last year’s In re Antor Media Corp opinion; these cases trace opposite edges of the enablement requirement for anticipation.

Inventor Steve Morsa applied for a patent on a computerized system and method for benefits matching in US application 09/832,440. The claimed invention basically provided a database of benefits – government assistance programs, discounts and special offers from private providers, and other freebies – and queried the database for benefits that matched the parameters of individual benefit-seekers. The claims dressed it up like this:

A method of generating a benefit result list . . ., comprising:
maintaining at least one database including a plurality of benefit listings;
receiving a benefit match request from a seeker, said request including said seeker’s criteria;
identifying those of said benefit listings having criteria which generate a match with said match request;
generating automatically a message to inform directly said seeker via [a] computer network of those said benefits which match said seeker’s criteria.

The USPTO rejected the claim as anticipated over a single piece of prior art called PMA, which was a press release describing a software program designed to match benefits with eligible consumers. Other claims were rejected as obvious over the PMA alone. Morsa attacked the press release (reproduced in its entirety here) as non-enabling and as not proper prior art due to a later publication date. The Board, however, observed that the enablement standard for a § 102 rejection was low enough to be met here and further dismissed Morsa’s approach on enablement as mere attorney argument, stating, “Appellant has not submitted declarations or affidavits to establish that the PMA reference was not enabled. The Appellant relies on arguments presented in the brief. Arguments in a brief cannot take the place of evidence. See In re Pearson.” The Board further credited the earlier publication date within the PMA publication itself in refuting Morsa’s argument of later publication.

The Federal Circuit took a rather different and interesting view of Morsa’s non-enablement argument. The court first acknowledged the rule of Antor, that prior art publications like PMA are presumed enabling; the Applicant bears the burden of proving a piece of prior art does not enable the subject matter of the claims. In the specific facts presented in Antor, expert declarations were insufficient to show non-enablement. But the Morsa court noted a gloss on this rule. Expert declaration and other hard evidence are not always required to rebut the presumption of enablement; rather, when a piece of prior art has enablement problems “on its face,” attorney argument alone must be substantively considered:

our decision in Antor cannot be read to require an applicant to submit affidavits or declarations to challenge the enablement of prior art references.

The presumption in Antor is a procedural one — designed to put the burden on the applicant in the first instance to challenge cited prior art; the PTO need not come forward with evidence of enablement before it may rely upon a prior art reference as grounds for a rejection. Once an applicant makes a non-frivolous argument that cited prior art is not enabling, however, the examiner must address that challenge. While an applicant must generally do more than state an unsupported belief that a reference is not enabling, and may proffer affidavits or declarations in support of his position, we see no reason to require such submissions in all cases. When a reference appears to not be enabling on its face, a challenge may be lodged without resort to expert assistance.

The panel found that Morsa had pointed out real reasons why the PMA was not enabling on its face – its brevity (117 words), its non-technical nature, and its reference to unexplained underlying operational features – and the USPTO failed to substantively address them. The panel declined to referee enablement on the merits for the first time on appeal, noting that an Examiner must first determine whether the teachings of a particular piece of prior art do, in fact, enable the claimed invention. The panel instead vacated the anticipation rejection as procedurally infirm and remanded for substantive consideration of Morsa’s argument.

The court did, however, affirm the Board’s determination of the earlier PMA publication date and further affirmed the Board’s obviousness rejections (on other claims) because the factual underpinnings of these determinations were supported by substantial evidence. Rejection for obviousness affirmed, rejection for anticipation on two other claims vacated and remanded.

Lessons: For overcoming anticipation by showing non-enablement, previous lessons from Antor supposed that “where a reference says something clearly ridiculous or self-contradictory, a lay/legal argument of non-enablement might work using the reference alone.” Morsa ends the guessing: yes, a ridiculously short and undetailed reference may be shown as non-enabling through argument alone, and the argument deserves a substantive response from the Examiner and Board. Mere attorney argument, however, won’t necessarily carry the substantive merits; indeed, Morsa may still lose when the Board substantively replies to his argument. Thus, the more conservative lessons from Antor ought to be followed where possible: try to get expert declaration or other hard evidence in the record to overcome anticipation due to non-enablement. This evidence should focus on the scope of the claimed invention and how that scope is not enabled, i.e., requires undue experimentation under the applicable Wands factors to practice, given the specific teachings of the allegedly-anticipatory reference alone.

From a more general standpoint, Morsa reflects that sometimes attorney argument alone is enough to traverse a prima facie rejection, even on a highly factual issue like anticipation and undue experimentation required of the prior art. So if an Examiner summarily waves off a traverse pointing out a deficiency of the prior art as “mere attorney argument,” consider citing Morsa if the deficiency is clear on the face of the reference. I hesitate to extend Morsa much beyond enablement, but I don’t see why mere attorney argument wouldn’t equally suffice in factual obviousness contexts like showing that a modification of the prior art is non-functional or changes operational principle if the prior art spells out that functionality exists only in its unmodified state. Really, this is not much different than arguing a lack of motivation to combine, the existence of teaching away, or that a proffered substitution is not between a known element performing a known function. And mere attorney argument on these issues is usually enough to trigger a substantive response from the Office.

Comment: I don’t find the court’s “on its face” requirement all that helpful for determining exactly when attorney argument alone is procedurally sufficient to merit a substantive response on a rejection. Whether a question of enablement – or any other deficiency – is apparent “on the face” of a publication is a fairly subjective inquiry and, moreover, one that is bound up with the ultimate substantive issue of whether the deficiency requires withdrawal or maintenance of the rejection. I think the Morsa court could have instead concluded that a substantive response from the Office is required whenever an Applicant’s argument is firmly based on evidence of record and not conjured out of thin air. For example, in Morsa the evidentiary backbone for the argument of non-enablement was the brevity and non-technical characteristics of the PMA reference of record. This approach might keep Applicants better focused on the record in their remarks and prevent Examiners from dismissing arguments whenever a 132 declaration is absent. I thought the above-quoted excerpt from Morsa was going in this direction, but it instead detoured to this “on its face” standard.

The Morsa court remanded the anticipation rejection without commenting much on the merits of the Applicant’s underlying non-enablement argument. But couldn’t the court have simply affirmed on evidence of record? The Board had other discussion in its decision as to how the PMA met the rather low § 102 enablement standard. Moreover, by affirming the 103 rejections over the PMA reference to other very similar claims, didn’t the court inherently find the PMA reference enabling for the claimed subject matter? After all “in order to render a claimed apparatus or method obvious, the prior art must enable one skilled in the art to make and use the apparatus or method.”

Lastly, sometimes defenders of § 101 rejections argue that subject matter eligibility works well as a “threshold” matter because it is an easy way to immediately kill claims on a legal issue without the need for burdensome examination and factual argument. Morsa’s claims look suspect under § 101 following Cybersource and Fort Properties (rightly or wrongly). By remanding for procedural considerations, doesn’t the court make a § 101 toss look like an appealing alternative in cases like these for administrative efficiency?

Case No. 2010-1426 (O’Malley, Bryson, Linn)

Appealing claim constructions to the Federal Circuit is something like gambling. Infringement damages typically dwarf costs of an appeal, so a 24% reversal rate on appealed claim constructions probably looks like a smart bet to would-be appellants. It is telling that parties familiar with gambling probabilities, like Aristocrat and IGT, are frequently on the court’s docket. The Aristocrat Tech. v. Int’l Game Tech. II opinion pays out with some good discussion of plain language treatment of claims and the role of extrinsic evidence like dictionaries and expert testimony. There are also lessons here for the relative value of apparatus and method claims based on one’s infringement target.

Aristocrat held US Patents 7,056,215 and 7,108,603, directed to methods of operating and playing networked slot machines. IGT had been internally testing Aristocrat-like machines, perhaps for functionality or design-around purposes, which Aristocrat alleged infringed the ‘215 and ‘603 methods. Merely testing the machines, however, fell outside the district court’s construction of the following claims, resulting in summary judgment of non-infringement:

1. In a network of gaming machines, each of said gaming machines having a user interface activatable by a player to affect game display, each of said gaming machines being capable of accepting different wager amounts made by the player, a method of randomly awarding one progressive prize . . ., comprising:
making a wager at a particular gaming machine in the network of gaming machines;
[several steps of playing and triggering progressive prize-winning];
identifying to the player said one progressive prize from said plurality of progressive prizes that has been won; and
awarding said one progressive prize from said plurality of progressive prizes that has been won.

Starting with the “awarding” limitation, Aristocrat argued that the step required only displaying a monetary amount, so that a tester would infringe the step simply by being shown what would have been won. Requiring actual legal entitlement to pass to a player was at odds with portions of the specification and would result in infringement being impossible in some jurisdictions where gambling awards were prohibited.

The panel saw no reason to depart from the plain meaning of “awarding,” which required actually doling out the prize money. To construe the term as requiring only an indication of the calculated award made the separate “identifying” step redundant and awkward. The fact that both steps made their object the same “said one progressive prize” meant they could not sensibly be doing anything other than identifying and separately awarding the prize. The court also noted that the specification equated “identifying” with displaying, which suggested “awarding” required something more. The court was not impressed with the argument that varying laws would make an “awarding” step that involved actual payment invalid in certain jurisdictions:

The critical aspect of the construction of the “awarding” step is the transfer of the right to a prize from the casino operator to the player—a transfer that not even Aristocrat alleges occurs during IGT’s testing of their equipment. That a particular state’s law may complicate a player’s ability to enforce collection of their prize does not affect the operation of the system claimed in the asserted patents or our construction of the “awarding” step.

Lastly on the construction of “awarding,” Aristocrat took issue with the district court’s use of a dictionary to define “awarding” to require a player gain legal entitlement in a prize. The panel saw no problem with this use of extrinsic evidence, because the dictionary was consistent with the intrinsic evidence and did not control the construction:

the district court simply used the provided definitions to inform its understanding of the “ordinary and customary” meaning of the word “award.” See Phillips, 415 F.3d at 1322-23 (“Dictionaries or comparable sources are often useful to assist in understanding the commonly understood meaning of words and have been used both by our court and the Supreme Court in claim interpretation.”); see also id. (“[J]udges are free to consult dictionaries . . . at any time in order to better understand the underlying technology and may also rely on dictionary definitions when construing claim terms, so long as the dictionary definition does not contradict any definition found in or ascertained by a reading of the patent documents.”)

On the “making a wager” limitation, the district court construed this term to require player action, such that direct infringement was lost when the independent player performed only the first step of the recited method. Aristocrat argued that the “making a wager” limitation was actually performed by the machine, noting that other claim steps explicitly recited “by the player,” whereas the wager-making step did not.

The Federal Circuit affirmed the district court’s construction here as well, looking mostly at the intrinsic record. Several portions of the abstract and description of the prior art equated playing – done by a player – with betting, suggesting the betting and its synonymous claim term “making a wager” were equally performed by a player. Moreover, Aristocrat made a change during prosecution from claim language that required only the slot machine to language that did not:

As originally drafted, the “making a wager” step made clear that “betting” is performed by the player and that the machine “allows” such betting: “allowing the player to bet a plurality of credits for a single play at a gaming machine in the bank of gaming machines.” J.A. 575 . . . Through an amendment, Aristocrat modified the above language as follows, resulting in the language that issued as the “making a wager” step: “allowing the player to bet a plurality of credits for a single play making a wager at a particular gaming machine in the bank network of gaming machines.” J.A. 629 (emphasis and strikethrough in original). By removing the reference to the actor that “allows” the betting, Aristocrat, at best, introduced ambiguity into the claim. The amendment did not clearly limit the claim to activity by the machine as opposed to the player.

Aristocrat lastly argued that its declaration by an expert named Crevelt, showing that “making a wager” was executed by the slot machine’s micro-processor, had been wrongly ignored. The courts, however, found the expert declaration conclusory. While the declaration stated that the slot machine had to process the bet, it did not explain why “making a wager” was limited to the machine’s processing. This amounted to nothing more than an unsupported assertion of claim interpretation by an expert and was dismissed by the court:

We agree with the district court’s characterization of Mr. Crevelt’s declaration and see nothing improper in discounting conclusory statements as “not useful.” See Phillips, 415 F.3d at 1318 (“[E]xtrinsic evidence in the form of expert testimony can be useful to a court for,” inter alia, the purpose of “establish[ing] that a particular term in the patent or the prior art has a particular meaning in the pertinent field. However, conclusory, unsupported assertions by experts as to the definition of a claim term are not useful to a court.”) (citations omitted). While Crevelt’s declaration explains, at length, how a gambling machine’s software “carr[ies] out or implement[s] a bet by executing computer code,” it fails to explain whether the patent’s use of “making a wager” refers to: (1) the steps performed by the gaming machine when it implements a bet; or (2) to the antecedent acts performed by the player when she places a bet. See J.A. 442-43. Instead, Crevelt’s declaration “assumes that ‘making wager’ means processing a bet” and then explains how the machine performs the processing.

. . .Aristocrat urges us to find that Phillips requires only that an expert explain “how” a person of ordinary skill in the art would understand the term, but that an expert need not explain “why” a person of ordinary skill in the art would understand a term in a particular manner. This argument ignores Phillips’s teaching that “conclusory, unsupported assertions by experts as to the definition of a claim term are not useful to a court.” 415 F.3d at 1318 (emphasis added). By failing to explain why a person of ordinary skill in the art would understand “making a wager” to describe the steps performed by the microprocessor in processing the bet, Crevelt failed to support his assertion that “making a wager,” as understood by a person of ordinary skill in art, “means transfer of credits from the credit meter to the bet meter by the game software.” J.A. 443. We find that the district court properly applied Phillips when it discounted Crevelt’s declaration.

In affirming these issues of claim construction, the court also affirmed that no direct infringement could be found against either the internal testing or actual use for gambling due to lack of command or control among completing actors under Akamai. The court did, however, remand on the issue of indirect infringement, suggesting a jury could find that IGT may have induced gamblers to complete the recited methods.

Lessons: Many times these lessons focus on achieving maximum subject matter breadth in claims. It’s important also not to lose sight of direct infringement in claim drafting and prosecution. Akamai loosened the requirements somewhat for direct infringement, permitting multiple actors to complete all steps of a method claim without defeating direct infringement, so long as a single actor, directs, controls, or would otherwise be vicariously liable for the acts of the multiple actors. But it’s still important to ask the question of “who’s going to perform this element?” for each element whenever drafting or amending claims. If you can’t identify a commercially relevant single party, or at least a single controlling party, for all claim aspects, consider redrafting the claim to instead include actions or parts originating from the single party. And perform this analysis even when amending claims in prosecution. In Aristocrat v. IGT II, it was really an amendment from “allowing the player to bet” to “making a wager” that changed the party performing the claim step away from the single party performing the remainder of the claim.

It is inherent in the above suggestions that a claim drafter know an applicant’s industry and even business model to identify single actors, and, moreover, to identify commercially-important single actors when writing claims for infringement. It always makes sense to direct at least some of the pre-drafting discussion during IP harvesting and disclosure toward these commercial realities to facilitate direct infringement. If a method’s novelty requires action among multiple independent actors, identification of this circumstance from the outset allows, at a minimum, better client counseling and buy-in this situation likely requiring indirect infringement. Even better, it may allow a client to structure contracts or actors in better direct infringing positions from the outset. Yet better still, a hard look at infringing parties from the outset may permit identification of a non-method – an apparatus or composition – whose use by any party will yield direct infringement. Indeed, in Aristocrat v. IGT II, had the claims been directed to slot machines configured to carry out the method, there likely would have been direct infringement by the gamblers and casinos (although perhaps still not the internal testers).

As for claim construction, Aristocrat v. IGT II reemphasizes the importance of the intrinsic record in determining claim scope. The court went again and again to the claim structure and specification to confirm the meaning of plain terms like “awarding” and “making a wager,” while eschewing inconsistent extrinsic declarations that failed to explain why, with support from factual sources in the record, a claim term should be interpreted in a particular manner. The case also reminds to be careful when defining actors in the specification – even in the background section and abstract. The court here readily looked to description of the background technology and abstract to see how claim terms were used. The Aristocrat v. IGT II panel was also comfortable looking to usage of words like “betting” and “playing” in example embodiments to determine the meaning of “making a wager”; ditto for “displaying” as to “identifying.” Thus, be mindful of potential limiting everywhere in an application. Consider adding in language that a single actor may be performing all identified steps, if possible, and avoid using claim terms and synonyms in connection with limited embodiments or actors anywhere in the description. Lastly, the case is yet another recent example of the court interpreting claims to avoid making any element redundant; be wary of any claim interpretation in litigation or prosecution that ignores a claim element or makes it redundant with a different claim element.

Comment: There have been a number of important cases captioned with “Aristocrat” and/or “IGT” from the Federal Circuit. This makes shorthand reference (“the Aristocrat opinion”) rather imprecise. Here are the full cites and issues for a few of them:

• Aristocrat Techs. Australia PTY Limited v. International Game Tech., 521 F.3d 1328 (Fed. Cir. 2008): means-plus-function claiming for programmed computers requires disclosure of algorithm in specification
• Aristocrat Techs. Australia PTY Limited v. International Game Tech. I, 543 F.3d 675 (Fed. Cir. 2008): improper revival is not an invalidity defense to infringement, as defined by 35 USC § 282
• Aristocrat Techs. Australia PTY Limited v. International Game Tech. II (Fed. Cir. 2013): this case
• IGT v. Alliance Gaming Corp, 702 F.3d 1338 (Fed. Cir. 2012): antitrust identification of competition within a relevant market.
• IGT v. Bally Gaming International, Inc., 659 F.3d 1109 (Fed. Cir. 2011): claim construction on “one”-to-one relationships, “upon the occurrence of” cause-and-effect relationships, and “predetermined” (blog post on this case here)

Case No. 2012-1265 (Lourie, Clevenger, Wallach) (nonprecedential)

Was it something I said? Statements made during prosecution are one of the scarier means by which a claim can later be narrowed in litigation. While courts typically rely on claim language and specification disclosure first in limiting claims, prosecution statements seem like both the most directly tied to the prosecutor and yet most subjective part of the intrinsic record. Smith & Nephew v. Arthrex is an illustration of safer distinguishing approach in prosecution that may avoid unduly narrow claim constructions in litigation. The case also shows why it is a good strategy to present at least one claim to the smallest aspect of a disclosure, despite a lower potential to rack up big infringement damages.

Smith & Nephew asserted US Patent 5,601,557 against Arthrex, accusing Arthrex of direct and induced infringement of the ‘557 patent with several of their bone anchors. Following a previous trip to the Federal Circuit, the district court granted judgment as a matter of law to Arthrex based on a narrow construction of the common connection term “lodging,” overturning a jury verdict of infringement. Particularly, the district court found that “lodging” as recited in the following claim required nearly permanent joining and stability for the lodged anchor so that it would withstand all forces of surgery:

1. A method for anchoring in bone a member and attached suture, comprising the steps of:
forming a hole in the bone;
attaching a suture to a member;
lodging the member within the hole by pressing the member with attached suture into the hole; and
attaching tissue to the suture so that the tissue is secured against the bone.

On appeal, the Federal Circuit found that the “lodging” step was not as limited as the district court had construed it. Rather, “lodging” here would permit small movements of the anchor and general instability to be be remedied later in surgery. In arriving at this construction, the panel looked at the prosecution amendment that added the “lodging” language to overcome prior art, as well as the prosecutor’s statements regarding the amendment. Here’s the entire transaction from the opinion:

“lodging” was only added to the claim language to overcome an anticipation rejection based on U.S. Patent 4,409,974 (“Freedland”). Freedland discloses connecting two sets of bone by using an anchoring device inserted into a hole drilled through the bone. That device was initially smaller than the diameter of the drilled hole, and it was not until the surgeon used pliers to deploy the arms of the device so that the anchor could remain in place. In overcoming that rejection, Hayhurst added “lodging” to the claims, stating:

This method [of lodging the member within the hole] is not found in the umbrella-like operation of the Freedland device. In order for the Freedland device to become lodged within a hole in a bone . . . it is necessary to apply opposing forces to the head 12 in order to swing the hinged arms 20 out of their collapsed position and into an extended position.

The Federal Circuit found these prosecution statements non-limiting. The distinguishing argument never mentioned anything about a degree of lodging or whether the lodging was permanent; rather, the amendment and explanation appeared directed to the mechanisms for achieving lodging – the prior art required a pinch to the anchor head whereas the claim language recited a press of the anchor into the bone. In the words of the court, “there is no mention that the claimed anchor must withstand all the forces of surgery. Instead, that statement appears to deal only with how the claimed anchor stays in place after being initially pressed into the hole in the bone.”

The panel also relied on the claim language and specification. Interestingly, the court reasoned that, because the claim was directed to just a small part of the complete surgery, the “lodging” would not have to be the full securing required to complete the surgery. The preamble and actual claim steps made this clear:

we see no support in the intrinsic record to require the “lodging” to “withstand[] all the forces of surgery.” Claim 1 of the ’557 patent is directed only to anchoring a suture in the bone, not to the full surgical operation. ’557 patent col. 11 ll. 2–9. The “lodging” step occurs just after “forming” the hole and “attaching” a suture, but prior to “attaching tissue.” Id. None of the other steps of the actual surgery that occur after that attachment of tissue to the anchor are claimed.

The specification also identified an initial lodging and then a separate “surgeon’s tug” that made the anchor permanently lodged to withstand all further forces, which also suggested that the initial insertion of the anchor could be somewhat loose.

Given the evidence of trial and the corrected construction of “lodging,” the Federal Circuit reversed the district court’s overturning the jury verdict. There was also an issue of whether Arthrex induced or contributed to infringement. This post does not address the issue further, but the panel split as to whether substantial evidence that Arthrex knew or was willfully blind as to whether the induced acts constituted patent infringement. Give this case a read for a good two-sided discussion of inducement following the 2011 Global-Tech Supreme Court case.

Lessons: The prosecution statements excerpted by the Smith & Nephew court are pretty well-executed. The applicant picked a single limitation – the one added by amendment – and focused on the teachings of the prior art – not the limitation – in distinguishing. This strategy of picking one claim aspect and avoiding characterizing the limits of that aspect has worked well in the past. If you look at the prosecution history, the ‘577 applicant actually did something else valuable – underlined the “by pressing. . .” language in order to make clear the argument was distinguishing how the anchor lodged differently from the prior art, not what “lodging” was limited to or to what degree “lodging” required.

As for claim strategy and structure, Smith & Nephew shows that it’s a good idea to include different claimsets directed to varying levels of vertical scope. Having an independent claim that focusses on the smallest inventive and workable unit of a disclosure can avoid importing other limitations that might be required of a larger combination. You can go after big damages in a separate independent claimset directed to the bigger device including all inventive aspects. Here, the ‘577 patentee successfully showed that “lodging” did not require the narrower construction of withstanding all forces throughout the entire surgical procedure, because the preamble, claim steps, and specification all reflected that the claims at issue dealt with only a small part of what would be the full procedure. In sum, don’t miss the forest for the trees, but don’t forget to claim both the forest and the (inventive) trees.

Comment: The claim construction looks about right to me. I could see a panel with a bit more agency for narrow constructions twisting the prosecution statements another way. For example, by comparing the very secure lodging of the prior art to the recited lodging step, and only distinguishing it by the way the lodging was achieved, the patentee might have been said to be admitting that “lodging” required the full surgical lodging but the prior art just achieved it in a different way. I’m glad the court didn’t go down that path, because even taking this interpretation as potentially correct, there were still multiple interpretations of the prosecution statements, which is supposed to bar prosecution estoppel.

Case No. 2012-1255 (Prost, Moore, O’Malley) (nonprecedential)

Breadth in claims can be achieved in a number of ways, both with the claim language itself and external tactics like claim differentiation or redefining a claim term more broadly in the specification. When trying to narrow a claim to avoid prior art, however, the primary approach is to simply focus on changing the claim language. In re Sutton includes a short discussion reminding that all sources of broadening, both within and outside of the specific claim language, should be taken into account when trying to narrow around the prior art. In Sutton, claim differentiation supported a broad construction of several claims in prosecution.

Sutton filed application 12/190,101 on a vinyl coating that did not require a plasticizer to preserve color and flexibility. All claims in the application were eventually rejected as obvious over a four-reference combination, which the Board twice affirmed. The structure and relationship among the claims was important, as paraphrased in the opinion:

7. A vinyl film subject to outdoor weathering, comprising a fluorescent dye and a terpolymer of ethylene—vinyl acetate—vinyl chloride that imparts color stability and flexibility to the film.

8. The film of claim 7, further comprising a plasticizer.

9. The film of claim 7, further comprising polyvinyl chloride.

30. The film of claim 7, wherein the film does not include a plasticizer.

The applicant argued before the Board and Federal Circuit that claims 7 and 9 could not cover a composition including a plasticizer under claim differentiation, because claim 8 explicitly recites the use of a plasticizer. Such a construction might not have been obvious, because the primary reference taught some embodiments that included a plasticizer.

This argument actually hurt the applicant before the court. The panel stated that, given the “comprising” transition, the addition of the plasticizer element in claim 8 did not exclude it from claims 7 and 9 under claim differentiation:

Claims 7 and 9 both use the transitional term “comprising” prior to reciting the required elements of the claimed composition. While the recited elements in claims 7 and 9 are necessary to form the claimed composition, by using the open-ended phrase “comprising,” the claims may include other unclaimed elements, such as a plasticizer.

. . .Sutton could have limited the scope of claims 7 and 9 by adding an explicit “no plasticizer” limitation, much like he did in claim 30. For whatever reason, Sutton did not choose to impose such a limitation on these claims.

In addition, claim 8 requires a generic plasticizer and claims 14 and 19 require different forms of plasticizers. These claims are narrower than claim 7 because they each require a particular type of plasticizer, while claims 7 and 9 simply permit the use of one.

Rather, claim differentiation worked the exact opposite way sought by the applicant here – because claim 30 did literally and only exclude plasticizers, claim 7 must permit plasticizers in its formulation:

Contrary to Sutton’s position, moreover, a straight-forward application of the doctrine of claim differentiation actually confirms, rather than undermines, this reading of the claims. Claim 30, which expressly excludes a plasticizer, would be superfluous were claims 7 and 9 read to exclude plasticizers.

The court noted that, in the alternative, claim 30 was found to be obvious over embodiments that did explicitly exclude plasticizers. So it would not have likely mattered if the claims had been limited as sought by the applicant. The court also refuted haphazard arguments by the applicant that the references were not combinable or would not have possessed the recited “color stability” would not have inherently existed in the combined references. The court accepted the Examiner’s and Board’s findings as supported by substantial evidence that there was motivation to combine all asserted references and that the references taught the use of a known color stabilizer.

Rejections affirmed.

Lessons: Claim differentiation is a valid basis to support a broad claim construction in prosecution, and an Examiner might fairly point to narrowing in dependent claims to interpret an independent claim term beyond the subject matter of the dependent claim. So watch out for dependent claims that undermine arguments against overbroad interpretations of independent claims in prosecution. If you have an independent claim reciting “element A” and a dependent claim narrowing element A to “A1,” an argument that the independent claim must be construed as limited to A1 to overcome the prior art disclosing “A2″ probably won’t work. It’s especially not going to work in a situation like in Sutton where a different dependent claim narrowed element A to “A2.” Instead, either amend your claim to capture the dependent claim having the unmet species or argue that dependent claim.

Now Marine Polymer comes to an essentially opposite conclusion on claim differentiation. There the en banc plurality narrowly construed independent claims to wholly exclude some dependent claims in prosecution while explicitly rejecting claim differentiation. But I read that case as dependent on very strong limiting language in the specification that forced the independent claim to be narrowly construed in prosecution despite claim differentiation. In such a situation, it’s probably best to cancel the dependent claims and argue the spec as controlling the construction if you need the limited specification construction for prior art avoidance purposes.

Comment: Sutton’s application and the Board opinions are not published (or, at least, I can’t find them on the Board’s terrible awful opinion search page). I’ve always wondered about the requirement that unpublished applications be held in confidence versus publication of tribunal opinions on unpublished applications. The confidentiality is often defeated, at least in substantive part, by the tribunal opinion, as would appear to be the case in Sutton. Is there authority that an applicant waives the confidentiality requirement by proceeding to appeal? I’m guess there is; can anyone help me out in the comments? Lastly, the court suggests the arguments on appeal here were not very well-presented or persuasive. Why no Rule 36 affirmance here for an unpublished application in a nonprecedential case?