Since the institution of inter parte review (IPR) proceedings under the Leahy-Smith America Invents Act in 2012, parallel proceedings before the Patent Trial and Appeals Board and district courts have become somewhat commonplace. Frequently, the former proceedings arise after a patent infringement defendant is served, as a way to accomplish one of the asserted goals of IPRs: reducing the costs of defending against such lawsuits. If the outcomes arise as expected, i.e., the PTAB invalidates some or all of the claims asserted before the district court, prevailing defendants can hope to recoup their attorneys’ fees and costs arising from the litigation but with less frequency from the IPR. However, prevailing in actions to recoup are not routine; indeed, the statute depends on the court finding them to be exceptional under 35 U.S.C. § 285. In Extremity Medical, LLC v. Nextremity Solutions, Inc. (nonprecedential), the Federal Circuit illustrates the difficulties that can arise in satisfying those statutory requirements for both types of proceedings.
The case arose involving Extremity Medical’s U.S. Patent No. 8,303,589, directed to orthopedic implant devices. After a series of accusatory letters from Extremity and denials by Nextremity, Extremity brought suit with a complaint alleging that a particular Nextremity product, the InCore Lapidus System, infringed claim 59 of the ‘589 patent. Within six months of being served, the PTAB instituted an IPR against the ‘589 patent and the District Court entered a stay upon the parties’ stipulation based on institution of the IPR.
Patent Challenger Nextremity asserted prior art references in support of its contentions that the asserted claims were invalid for obviousness, which references included U.S. Patent Nos. 4,622,959 (“Marcus”) and 6,579,293 (“Chandran”). Extremity for its part submitted motions to amend asserted claim 59 without further submitting arguments in support of its patent claims’ continuing validity, to which the PTAB responded by a determination (in a Final Written Decision) under the preponderance of the evidence that claim 59 was invalid for obviousness using the two asserted references and a third unrecited reference. Additionally, the PTAB found that Extremities’ proposed amended claims were also invalid for obviousness, the Board faulting IPR counsel for not providing arguments in support of these amendments. This FWD being issued, the District Court lifted the stay on proceedings before it and dismissed the action with prejudice. Nextremity then moved for attorneys’ fees and costs relating to actions before both the PTAB and the District Court. The District Court found that the circumstances surrounding the litigation fulfilled these requirements and awarded Nextremity $52,573 in fees and costs but denied its demand for $343,660.86 for the IPR proceedings. This appeal arose over Nextremity’s attempt to recoup its fees and costs for these IPR proceedings and Extremity’s cross-appeal challenging the District Court award to Nextremity for the fees and costs for the litigation.
The Federal Circuit affirmed on both decisions, in an opinion by Judge Lourie joined by Chief Judge Moore and Judge Cunningham. With regard to Nextremity’s appeal, the panel opinion noted that the District Court relied on Dragon Intell. Prop. LLC v. DISH Network L.L.C., 101 F.4th 1366, 1371 (Fed. Cir. 2024) (citing Waner v. Ford Motor Co., 331 F.3d 851, 857 (Fed. Cir. 2003)), which held that a party could not collect attorneys’ fees and costs for a “voluntarily undertaken parallel IPR proceeding[]” under § 285. That Court’s rationale was that the IPR proceedings were not “part and parcel” of the district court litigation but rather were “strategically pursued . . . in lieu of litigating invalidity before the district court.” Additional precedent supporting this decision characterized IPRs as not being “cases” under § 285 under Amneal Pharms. LLC v. Almirall, LLC, 960 F.3d 1368, 1371–72 (Fed. Cir. 2020), and further did not have the benefit of district court proceedings wherein the court is “particularly well positioned to determine whether a case before it is exceptional because it ‘lives with the case over a prolonged period of time,’” quoting Highmark Inc. v. Allcare Health Mgmt. Sys., Inc., 572 U.S. 559, 564 (2014). Nothing in the proceedings below compelled Nextremity to pursue its claims before the PTAB in an IPR in the panel’s opinion.
The panel rejected Nextremity’s argument relying on Sullivan v. Hudson, 490 U.S. 877 (1989), that cases can be deemed exceptional under circumstances “where administrative proceedings are intimately tied to the resolution of the judicial action and necessary to the attainment of the results Congress sought to promote by providing for fees, [wherein] they should be considered part and parcel of the action for which fees may be awarded.” That is not the case here, according to the opinion, and the quote (according to the panel) has been taken out of context by Nextremity. In Sullivan, the administrative proceedings at issue were mandatory (and on remand from a district court) and the resolution of the complaint depended on the outcome of the proceedings (although the parallels with the proceedings here are apparent). But that is not this case, according to the Federal Circuit. Finally, the Court rejected Nextremity’s argument that the costs from IPR proceedings were part of a “holistic, equitable approach” that requires consideration of the totality of the circumstances regarding exceptionality.
Turning to Extremity’s cross-appeal, the opinion enunciated the proper standard for a district court to apply under § 285 is whether the case “stands out from others with respect to the substantive strength of a party’s litigating position . . . or the unreasonable manner in which the case was litigated,” citing Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U.S. 545, 554 (2014). This determination is made on a case-by-case basis under the totality of the circumstances and reviewed for an abuse of discretion. Here, the bases supporting the District Court’s finding this case exceptional in favor of Nextremity included that Extremity “conducted no prelitigation investigation, even though it was on notice of the Marcus and Chandran prior art; (2) did not defend claim 59 before the Board; and (3) put forth no substantive arguments in favor of its litigation position throughout the district court case.” The Federal Circuit rejected Extremity’s arguments that the District Court had shifted the burden of proof and drew adverse inferences against Extremity, asserting that Extremity failed to rebut Nextremity’s evidence supporting its claim for attorneys’ fees and costs. The panel also disagreed with Extremity regarding the District Court’s weighing of the evidence, finding that the District Court acted within the scope of its discretion in combining Nextremity’s pre-suit letter and Extremity’s lack of pre-suit investigation (“[d]espite [its] awareness of relevant prior art years before commencing the suit”) and failure to make its validity case before the District Court. “Viewed together,” the opinion states, “the record reflects more than ordinary litigation weakness; it supports the district court’s conclusion that Extremity’s litigating position ‘stands out’ from typical patent disputes in which parties advance and defend colorable validity theories.” Accordingly, the Federal Circuit affirmed the District Court finding awarding attorneys’ fees and costs to Nextremity.
For the fifth consecutive year, we have surveyed how the Patent Trial and Appeal Board (PTAB) decides appeals of § 101 rejections from examiners. Readers of the previous installments will know the drill by now. Applicants who appeal a subject matter eligibility rejection to the PTAB have historically faced odds somewhere between long and laughable. The affirmance rate was 87.1% in 2021, 88.4% in 2022, 91% in 2023, and 88.6% in 2024.[i] Against the PTAB’s overall affirmance rate of roughly 60% across all grounds of rejection,[ii] these numbers painted a picture of an institution that, whatever the merits of any individual appeal, treated ineligibility as close to a foregone conclusion. We attributed this less to the weakness of applicants’ inventions and more to the Alice Corp. v. CLS Bank Int’l framework itself, a test so malleable that decision-makers could reach nearly any result while appearing to follow precedent, and one that the PTAB near-consistently resolved against applicants.
This year, for the first time, we have something resembling good news. On September 18, 2025, John A. Squires was confirmed as Director of the USPTO, and he wasted no time signaling that the Office’s approach to § 101 would change. Barely a week into his tenure, Squires wrote an Appeals Review Panel decision vacating a PTAB panel’s sua sponte § 101 rejection of a machine learning application in Ex parte Desjardins, chiding the PTAB for evaluating the claims at too high a level of generality and reminding it that §§ 102, 103, and 112 are the traditional and appropriate tools for limiting patent scope.[iii] The decision was designated precedential on November 4, 2025, and its reasoning has since been incorporated into the MPEP.[iv] PTAB panels took notice, with post-Desjardins decisions such as Ex parte Carmody reversing § 101 rejections in a manner that tracks the new guidance.[v]
So, has the fever finally broken? The answer, based on the 2025 data, is a qualified “somewhat.” The overall affirmance rate for 2025 was 84.9%, a modest but measurable decline from prior years, with most of the movement concentrated in decisions issued after Squires took office. Indeed, broken out by quarter, we found that the affirmance rate dropped to 75.4% in Q4 of 2025, after being 88.3% through Q1, Q2, and Q3.
To be clear, an applicant appealing a § 101 rejection still loses far more often than not, and one precedential decision cannot untangle a decade of Alice-induced confusion. But after four years of reporting that the situation was bad and nothing was changing, we can at least report that something has. Whether that something amounts to a long-term course correction or a blip remains to be seen.
As was the case in previous years, we applied a strict search methodology on the PTAB’s decision portal to identify qualifying § 101 appeals. Particularly, we specified the following criteria: decision dates between January 1, 2025 and December 31, 2025, a proceeding type of “appeal”, a decision type of “decision”, and an issue type of “101”. These search results were overinclusive and thus required filtering and a degree of manual review.
We only included decisions in which the applicant appealed an examiner’s Alice-based § 101 rejection and the PTAB ruled on this basis of appeal. To simplify the analysis, we considered a decision to be an “affirmance” of the examiner’s § 101 rejection if the PTAB held at least one claim invalid under § 101. Excluded cases were those in which: (i) the appeal was not of a § 101 issue but the PTAB set forth a new ground of rejection based on § 101, and (ii) the appeal was of a § 101 issue but the PTAB did not decide the case on § 101 grounds (e.g., the examiner withdrew the § 101 rejections after the appeal was filed). This gave us 977 qualifying appeals, of which 829 resulted in affirmance on § 101 grounds, yielding the aforementioned affirmance rate of 84.9%.[vi]
In order to determine the Squires Effect, we broke this data out by month.
As noted above, until Squires’ appointment, the 2025 data tracked the rates we had found in 2021-2024. But in October through December of 2025, the rates dropped. This suggests that the new Director’s policy change had an immediate, if limited, impact. To keep things in perspective and assuming the Q4 rates hold steady, an applicant appealing a § 101 rejection has about the same chance of success as correctly calling two coin flips in a row. The house still wins – and wins often. But for the first time since we began tracking these numbers, the house’s edge has visibly narrowed, and that alone is worth noting.
The data also provided additional insights. For example, when broken out by technology center, we found a pattern consistent with previous years – that the vast majority of appeals and affirmances come from TC2100, TC3600, and TC3700.
Having said that, the affirmance rates in all three of these technology centers dropped off in comparison to previous years.[vii] For example, TC3600 and TC3700 had hovered above 95%, while TC2100 had fluctuated between 75% and 86%. The drops in 2025 are notable, and even more so when considering just the Q4 data – TC2100 was 48.3%, TC3600 was 86.8%, and TC3700 was 73.1%.
We also considered situations in which the PTAB came up with its own grounds of rejection under § 101. This can occur when the PTAB reverses the examiner’s grounds of rejection and replaces it with their own, or when the examiner did not reject the claims under § 101 but the PTAB issued a sua sponte rejection. We were curious regarding the Squires Effect on this metric, as Squires wrote in Desjardins: “However, it is with this view that the panel’s sua sponte action is most troubling, as it eschewed the clear teachings of Enfish, and instead substituted only a cursory analysis that ignored this well-settled precedent. Panels should treat such precedent with more care, especially when acting sua sponte.”
Indeed, we found that this chiding seemed to have the desired result. While there were relatively few new grounds of rejection under § 101 in the data (at most 9 in any month), the quarterly numbers dropped from 9, 14, and 13 in Q1, Q2, and Q3, respectively, to 5 in Q4. The data here is too sparse to draw firm conclusions, but the early trend is promising.
Finally, 2025 had 4 utility rejections, more than we have seen in any previous year. These decisions are not reflected in the data above (as there was no Alice analysis). One such decision adopted the Examiner’s reasoning of “because the device, as disclosed, would violate Newton’s Laws of motion as well as fundamental Conservation of Momentum principles, the device is incapable of a specific, credible, and substantial utility as required by 35 U.S.C. § 101.”[viii] We take no position on this matter, aside from noting that the decision serves as a reminder that § 101 was intended to be a coarse filter to reject or invalidate clearly non-statutory inventions. The Alice overreach is judge-made law incongruously grafted onto the legislation.
So where does this leave us? After four years of documenting a tribunal that affirmed § 101 rejections with reliability akin to that of a sunrise, we can finally report a wobble in the orbit. The Squires Effect seems to be measurable, if modest in the grand scheme of things. An applicant appealing an eligibility rejection in late 2025 still lost three times out of four, which is nobody’s definition of a fair fight. But in a corner of patent law where the status quo has survived a decade of criticism from judges, scholars, and at least one long series of articles from frustrated practitioners at a certain Chicago-based firm, a ten-point drop in a single quarter qualifies as notable activity. Whether this is the beginning of a genuine correction or merely a brief administrative mood swing will depend on what the Director does next, what the PTAB internalizes, and whether anyone in Congress remembers that § 101 is their statute to fix.
[1] Walter DeGroft is a 2026 summer associate at MBHB. In the Fall, he will begin his final year of law school at Chicago-Kent.
[v]Ex parte Carmody, Appeal 2025-002843 (PTAB Dec. 30, 2025).
[vi] This total of 977 decisions is the highest by far of any of the four previous years in which we have conducted this study. The previous record was 708 decisions in 2021.
[vii] The table sums to 971, rather than 977, due to intentional omissions of the “experimental” technology centers, such as 3900, RD00, OPQA, OPT, and OPET. All decisions from these technology centers resulted in affirmance.
[viii]Ex parte Wood, Appeal 2023-004154 (PTAB Jun. 2, 2025).
In a nonprecedential opinion, the Federal Circuit affirmed a decision by the Patent Trial and Appeal Board that the claim challenged in an ex parte reexamination was invalid for anticipation in In re Magnolia Medical Technologies, Inc.
The appeal involved the ex parte reexamination of U.S. Patent No. 10,039,483, wherein a third party requestor asserted that claim 1 was anticipated by U.S. Patent No. 6,013,037 disclosing a syringe that minimizes contamination or hemolysis in the collection of multiple blood samples. The challenged claim was directed to a blood sequestration device and a specific configuration thereof; independent claim 1 was set forth in the opinion:
1. A blood sequestration device, comprising:
a housing having an inlet port configured to be fluidically coupled to a patient and an outlet port configured to be fluidically coupled to a sample reservoir;
a fluid reservoir disposed in the housing and at least partially defined by a seal member, the fluid reservoir configured to receive an initial volume of blood withdrawn from the patient (this limitation was referenced in the opinion as the “fluid reservoir limitation”); and
a vent disposed in the housing and configured to allow air to exit the housing as blood enters the fluid reservoir,
the blood sequestration device configured to allow the initial volume of blood to flow from the inlet port to the fluid reservoir,
the blood sequestration device further configured to allow a subsequent volume of blood to flow from the inlet port toward the outlet port via a sampling flow path, thereby bypassing the fluid reservoir and the initial volume of blood sequestered therein (this limitation was referenced in the opinion as the “bypass”).
(wherein italicized claim language was relevant to the issues before the Court on appeal). The advantage of this device was that it reduced contamination from dermally residing microbes in externally administered blood sources. The Board affirmed on appeal the Examiner’s rejection that the ‘483 patent claim 1 was anticipated. In so doing, the Board found that the ‘037 patent disclosed “a . . . first [blood] sample is contained within the fluid chamber 30” which tracked the fluid reservoir limitation recited in claim 1 of the ‘483 patent. Further, the Board found that the ‘037 patent discloses that the “fluid chamber 30 may be substantially separated from the second and any subsequent samples drawn,” which the Board considered to track the bypass limitation in the ‘483 patent claim. This appeal to the Federal Circuit followed.
In a per curiam opinion by a panel of Circuit Judges Lourie and Prost and Judge Arun Subramanian, District Judge for the Southern District of New York sitting by designation, the PTAB decision was affirmed. The panel’s decision was straightforward, the per curiam opinion asserted that Magnolia’s arguments on appeal were unpersuasive.
Regarding the fluid reservoir limitation, the panel considered Magnolia’s argument to be “built upon multiple premises,” explaining that:
[T]he phrase “initial volume of blood” as used in the fluid reservoir limitation should be construed to mean “the first volume of blood after venipuncture,” i.e., the insertion of a needle into the vein. . . . Magnolia then contends that Brannon’s fluid chamber 30 must be configured to be devoid of any air at the time of venipuncture to avoid the safety risk of introducing an air embolism into the patient [(based on an expert declaration)]. Accordingly, Magnolia asserts that the required configuration of fluid chamber 30 cannot meet the “an initial volume of blood from the patient” requirement of the reservoir limitation because it does not contain blood at the time of venipuncture.
However, the panel found nothing in the text of claim 1 of the ‘483 patent that required the limitation relating to a fluid reservoir be met at the time of venipuncture and that whether or not the limitation is met at the time of venipuncture does not affect its analysis regarding anticipation, citing Net MoneyIN, Inc. v. VeriSign, Inc., 545 F.3d 1359, 1370 (Fed. Cir. 2008). In addition, the panel did not find persuasive Magnolia’s arguments regarding “purported” safety concerns because one embodiment of the invention claimed in claim 1 of the’438 patent operates “in a materially similar manner” to the supposed unsafe operation of the device disclosed in the ‘037 patent. In the Court’s opinion “[m]ultiple premises upon which Magnolia’s argument are unsound, and thus the argument fails.”
Magnolia also argued that the ‘037 patent did not meet the bypass limitation because that patent disclosed the operation of the claimed syringe device having an “initial flash” of blood. This was contrary to the ‘483 patent claims requirement that the initial volume of blood was “sequestered” in the fluid reservoir for which there was no disclosure in the ‘037 patent. In the Court’s opinion, the ‘483 patent claim does not require that the “full amount” of the blood collected initially be sequestered, only “an initial volume” thereof (emphases in opinion). This distinction rendered immaterial that the ‘037 patent required the initial portion ends up in the collected blood sample and the panel did not recognize any temporal differences sufficient to provide a distinction between the claimed blood sequestration device and the prior art.
The readiness with which the Board and the Federal Circuit found that that challenged claims of the ‘483 patent were anticipated by the ‘037 patent is perhaps partially explained by recognizing that upon review of the ‘463 patent it can be seen that the ‘037 patent was not cited by the patent Examiner nor disclosed in an information disclosure statement by the applicants. Thus, the Office did not have a chance to consider the teachings of the ‘037 patent prior to the ex parte review that brought the patent to the Office’s attention for the first time, resulting in invalidation on anticipation grounds.
More than a decade after Alice Corp. v. CLS Bank Int’l, the two-step framework for patent eligibility under 35 U.S.C. § 101 remains as contentious as ever. Courts, commentators, and the U.S. Patent and Trademark Office have all acknowledged the difficulty of applying a test that turns on undefined terms such as “abstract idea,” “directed to,” and “significantly more.” But within this legal wreckage, a specific and recurring failure mode deserves closer attention than it has received. Particularly, how a challenger can readily neutralize a patentee’s otherwise strong eligibility argument for a software invention (i.e., a claimed technical improvement described as such in the specification) by asserting in a conclusory fashion that the improvement is to an abstract idea recited by the claim.
The sleight of hand works like this. The patentee points to Enfish, LLC v. Microsoft Corp. and its progeny, argues that the claims are directed to an improvement in computer functionality, and sets forth the specification’s disclosure of the technical problem and its solution. The challenger, rather than disputing that the claims recite an improvement at all, responds that whatever improvement exists is merely an improvement an underlying abstract idea. Invoking SAP America, Inc. v. InvestPic, LLC and BSG Tech LLC v. BuySeasons, Inc., for example, the challenger merely states that an advance an abstract idea cannot confer eligibility no matter how innovative the claimed invention. With that single unreasoned characterization, the patentee’s entire pro-eligibility argument often evaporates.
In many cases, notably those involving software inventions that reduce computational resource usage, this characterization is analytically improper. A claimed invention that reduces processor cycles, memory consumption, bandwidth, or latency is directed to a change in the physical operation of a machine. Calling that change an improvement to an abstract idea does not make it one. Yet the current § 101 framework allows the assertion to succeed with remarkably little scrutiny, both in the courts and at the USPTO.
The Federal Circuit has repeatedly held that improvements to computer functionality are not abstract. In Enfish, the Federal Circuit found claims to a self-referential database table eligible because they were “directed to a specific improvement to the way computers operate.” The Court cautioned against describing claims at “a high level of abstraction and untethered from the language of the claims,” warning that doing so would render the “directed to” inquiry meaningless because “all inventions at some level embody, use, reflect, rest upon, or apply laws of nature, natural phenomena, or abstract ideas.”
More recent decisions have characterized the pro-eligibility argument as having two parts – the improvement must be described in the specification, and the claims must reflect it. In Contour IP Holding, LLC v. GoPro, Inc., the Court reversed a summary judgment of ineligibility where the claims recited a point-of-view camera generating high- and low-quality video streams in parallel, with the lower-quality stream enabling real-time wireless preview on a remote device. The specification described the technological problem of streaming video from a resource-constrained camera, and the claims recited the specific means of solving it; that combination placed the claims on the Enfish side of the line. Conversely, in Recentive Analytics, Inc. v. Fox Corp., the Court held claims applying generic machine learning to television scheduling ineligible precisely because neither the claims nor the specification explained how any improvement to the underlying technology was accomplished. And in U.S. Patent No. 7,679,637 LLC v. Google LLC, No. 24-1520 (Fed. Cir. Jan. 22, 2026), the Court stated that while the eligibility inquiry focuses on the claims, “the claim itself need not explicitly recite the improvement,” and endorsed looking to the written description to understand the problem addressed and whether a technological improvement is embodied in the claims. Taken together, these decisions confirm that the specification’s technical disclosure is not window dressing, but rather evidence the challenger or tribunal is expected to consult.
The USPTO codified this line of authority in the 2019 Revised Patent Subject Matter Eligibility Guidance, now reflected in M.P.E.P. § 2106.04(d). Under Step 2A, Prong Two, a claim that integrates a judicial exception into a practical application is not directed to that exception. The first and most prominent example of such integration is “an improvement in the functioning of a computer, or an improvement to other technology or technical field.” M.P.E.P. § 2106.04(d)(1) and § 2106.05(a) instruct examiners to evaluate the specification to determine whether it describes a technical problem and explains how the claimed invention solves it. An August 2025 memorandum to the examining corps reinforced the point, reminding examiners that the claim need not explicitly recite the improvement described in the specification and directing them to consult the specification and confirm that the claim reflects the disclosed improvement.
On paper, then, one with a genuine technical improvement described in the specification and represented in the claim should be well positioned. In practice, the picture is far murkier.
The countervailing argument used by challengers traces back to two 2018 decisions. In SAP America, the court held claims to improved statistical analysis of investment data ineligible, stating that even assuming the techniques were “groundbreaking,” a pioneering abstract idea is not enough. In BSG Tech, the Court held that “a claimed invention’s use of the ineligible concept to which it is directed cannot supply the inventive concept,” and that an alleged advance that lies entirely in the realm of abstract ideas cannot confer eligibility.
Both holdings are defensible on their facts. The claims in SAP America were directed to mathematical analysis of financial information with the results displayed, and the claims in BSG Tech recited a method of indexing information in a database using conventional structures. Neither specification credibly described an improvement to the operation of a computer.
The problem is what happened next. These cases handed challengers and examiners a template that can be deployed against any software claim, regardless of what the specification actually discloses. In district court briefing, the pattern is now routine. The defendant frames the abstract idea broadly, characterizes every claimed benefit as flowing from that idea, and cites SAP America and BSG Tech for the proposition that such benefits are irrelevant. At the USPTO, the corresponding move appears in countless Office actions as a single conclusory sentence, something to the effect of “the alleged improvement is an improvement to the abstract idea itself, not to the functioning of a computer or to another technology.”
Note the asymmetry. To establish a technical improvement, the patentee must engage with the claim language, the specification, and often expert testimony explaining the state of the art and the nature of the advance. To defeat that showing, the challenger needs only a characterization. No claim construction, no engagement with the written description, no explanation of why a reduction in memory usage or processing time is not a change in how the machine operates. The characterization (often a mischaracterization) does all the work, and decision makers too frequently accept it at face value.
A pronounced flaw in the “improvement to the abstract idea” argument is its circularity, and this flaw is most visible in cases involving computational resource efficiency. Consider a claim to a data structure, encoding scheme, or scheduling technique that measurably reduces processor load, memory footprint, or network bandwidth. These are physical, measurable phenomena. A processor that executes fewer instructions consumes less power and completes tasks sooner. A memory system that stores a compressed representation of data uses less silicon. Reducing network utilization increases throughput and reduces delay for all communications carried by that network.
These effects occur in a machine, not in the mind of a mathematician. The Federal Circuit recognized as much in Enfish, where the claimed self-referential table achieved “increased flexibility, faster search times, and smaller memory requirements,” and in Uniloc USA, Inc. v. LG Electronics USA, Inc., where a reduction in communication latency was held to be a patent-eligible improvement to computer functionality. Similarly, in Cellspin Soft, Inc. v. Fitbit, Inc., allegations that the claimed arrangement reduced the hardware and power requirements of the capturing device were credited.
Now observe how the challenger’s argument operates against such a claim. First, the challenger defines the abstract idea at a level of generality high enough to encompass the improvement. If the claim reduces memory usage through a novel indexing arrangement, the abstract idea becomes “organizing and storing data.” If the claim reduces bandwidth through a selective transmission protocol, the abstract idea becomes “sending only necessary information.” Second, having drawn the abstract idea to swallow the technical contribution, the challenger points out that the claimed benefit flows from that contribution, and therefore from the abstract idea. Third, the challenger concludes that under BSG Tech or similar authority that the improvement “lies entirely in the realm of abstract ideas.”
This reasoning is entirely circular. The conclusion that the improvement is abstract depends on the premise that the abstract idea includes the improvement, and that premise was constructed by the challenger for exactly that purpose. Any invention can be dissolved this way. The claims in Diamond v. Diehr improve upon “the abstract idea of curing rubber at the right time.” The claims in McRO improve upon “the abstract idea of matching mouth shapes to sounds.” The claims in Ancora improve upon “the abstract idea of verifying a license.” The Federal Circuit warned against precisely this maneuver in Enfish and again in Thales Visionix Inc. v. United States, yet the “improvement to the abstract idea” formulation invites decision makers to commit the error while appearing to apply binding precedent.
The tell, in resource-efficiency cases, is that the claimed benefit is verifiable on hardware. One can run the prior art technique and the claimed technique on the same machine and measure the difference in cycles, watts, bytes, or time. Where a computer merely executes an unchanged process faster than a human could, as in Bancorp Services, L.L.C. v. Sun Life Assurance Co. of Canada, the improvement is attributable to the computer, not the invention. In these situations, eligibility is properly denied so long as the process was not reasonably performable by a human in the first place.[1] When the specification describes, and the claims capture, a change to the physical behavior of a computing system that results in a technical improvement, the SAP America line of cases simply does not fit. That the improvement can be described in abstract terms proves nothing, because as Enfish reminds us, every invention relies on underlying judicial exceptions.
If the circularity problem is troubling in litigation, where the patentee at least has briefings and oral arguments to expose it, it is worse in prosecution. Examiners operating under time constraints frequently dispose of an applicant’s eligibility arguments with the conclusory sentence described above, without addressing the specification’s description of the technical problem or explaining why the asserted benefit is not technical. This practice sits uneasily with M.P.E.P. § 2106.04(d)(1), which directs the examiner to evaluate whether the specification sets forth an improvement and whether the claim reflects it. A bare assertion that the improvement is to the abstract idea evaluates nothing.
PTAB review of such rejections has been inconsistent, with some panels demanding that examiners engage with the disclosed improvement and others affirming on the same conclusory rationale. The practical consequence is perverse. Applicants can respond by overloading claims with hardware recitations and implementation minutiae that add nothing inventive but make the claims look less abstract. Claim quality degrades, prosecution costs rise, and the resulting patents protect the invention less faithfully than the original claims would have.
Three corrective principles would go a long way. First, the party asserting that an improvement encompasses the abstract idea should bear the burden of engaging with the specification and articulating, with specificity, why the asserted benefit is not a technical one. A characterization unsupported by analysis should carry no weight.
Second, measurable reductions in computational resource usage should be treated as presumptively technical. This is not a radical proposition, as it follows directly from Enfish, Uniloc, KPN, and Cellspin. A challenger remains free to show that the claims fail to capture the improvement or that the specification’s assertions are implausible, but the physical nature of the benefit should not be in serious dispute.
Third, decision-makers should heed Diehr‘s instruction to consider the claim as a whole, rather than first filtering out the allegedly abstract components and then asking what remains to be improved. The filtering approach guarantees the circular result, because the improvement is always found among the components that were filtered out.
The Alice framework has always tolerated a certain amount of rhetorical gamesmanship, but the “improvement to the abstract idea” gambit is a particularly corrosive example. It targets the one doctrine, the Enfish technical-improvement line, that gives software patentees a principled path to eligibility. Until courts and the USPTO require those who invoke SAP America, BSG Tech, and the like to actually demonstrate that a claimed improvement is abstract, rather than merely to assert it, inventors of genuinely technical advances, including those that make computers demonstrably faster, smaller, and more efficient, will remain exposed to a one-sentence dismissal of their strongest argument.
Thus, the irony is complete. A doctrine allegedly created to weed out patents that claim inventions too abstractly is now enforced through arguments that reason too abstractly.
[1] Put another way, a resource-intensive task like training a large language model is not something that a human ever has performed or even could perform. Therefore, it is deeply reliant on the computing hardware on which it executes, without which there would be no model. Alternatively, a computer performing an unchanged human process is almost certainly to fail as obvious under § 103, so why bother using the vague and problematic analysis of § 101 for this purpose?
The Federal Circuit affirmed a District Court determination that method claims reciting administration of a modification of an established antibiotic by adding magnesium to the composition were infringed and not invalid in Melinta Therapeutics, LLC v. Nexus Pharmaceuticals, Inc.
The case arose as ANDA litigation over Nexus Pharma’s generic version of Melinta’s antibiotic formulations comprising magnesium cations used in a claimed method for treating bacterial infections that reduced hemolysis of red blood cells at the injection site, compared with such methods using antibiotics without the addition of magnesium, as claimed in U.S. Patent Nos. 9,084,802 and 9,278,105. Claim 1 of the ‘805 patent and claim 1 of the ‘105 patent are reproduced in the opinion as being representative:
The ‘802 patent:
1. A method of treating a bacterial infection in a subject, wherein the method consists of:
administering a therapeutically effective amount of a composition to a subject in need thereof via an intravenous route of administration,
wherein the composition consists of an aqueous solution consisting of minocycline or a salt thereof, a salt that comprises a magnesium cation, and a base,
wherein the molar ratio of magnesium cation to minocycline is greater than about 4:1, and
wherein the composition has a pH that is no less than 4 and no greater than 6,
whereby injection site hemolysis of red blood cells is reduced relative to intravenous administration of a composition that does not include magnesium.
The ‘105 patent:
1. A method of treating a bacterial infection in a subject, wherein the method comprises administering a therapeutically effective amount of a composition to a subject in need thereof via an intravenous route of administration, wherein the composition comprises an aqueous solution of a 7-dimethylamino-tetracycline antibiotic and a magnesium cation, wherein the molar ratio of magnesium cation to 7-dimethylamino-tetracycline antibiotic is greater than 3:1 and wherein the solution does not comprise a pharmaceutically acceptable oil, has a pH greater than 4 and less than 7, and has an osmolality less than about 500 mOsmol/kg.
Melinta’s product, Minocin, is an aqueous solution of minocycline and magnesium and, as the opinion notes, had been marketed since 1972 but had been reformulated to include magnesium, as well as having an increase pH and a reduced six of the administered dose.
The District Court found that Nexus Pharma’s generic formulation would infringe the patented method claims of the ‘802 and ‘105 patents and issued a permanent injunction. The District Court also held that Nexus had failed to establish invalidity of the claims on “obviousness, indefiniteness, inadequate description, or lack of enablement” grounds. This appeal followed.
The Federal Circuit affirmed with regard to the asserted claims of the ‘802 patent and did not reach the issues raised for the ‘105 patent, in an opinion by Judge Bryson, joined by Judges Prost and Reyna. With regard to the District Court’s findings concerning the ‘802 patent Nexus raised two arguments on appeal. First, Nexus argued that the District Court erred in construction of the terms “composition” and “administering.” Regarding the “composition” term, District Court construed the term to mean (as recited in the claim) the combination of minocycline, magnesium, and a base prior to the addition of the diluent needed to produce the formulation (the “admixture” as recited in the claim) actually administered to a patient in need thereof. The construction of the term “administering” by the District Court referred to the diluted composition given to the patient.
Nexus argued that the phrase as a whole in the claim (“administering . . . a composition”) required the administered composition not to include a diluent, because claim 1 recites that the composition “consists of” the three listed named ingredients and did not include a diluent (“wherein the composition consists of an aqueous solution consisting of minocycline or a salt thereof, a salt that comprises a magnesium cation, and a base”). Because Nexus Pharma’s composition does contain a diluent administration thereof is not infringing according to their argument. Melinta argued that the term “composition” should be construed to the “”concentrated, reconstituted solution prior to further dilution” and that what was administered was the composition after the diluent was added.
The District Court construed these terms consistent with Melinta’s argument and the Federal Circuit affirmed based on the intrinsic and extrinsic evidence. This included a description in the specification that the three-component composition was diluted prior to administration; the testimony of three expert witnesses to the same effect; and the prescribing information that instructs reconstitution and further dilution. Nexus Pharma’s proposed construction was inconsistent with this evidence, according to the opinion, and a person having ordinary skill in the art would know adding a diluent to the composition was necessary. The panel affirmed the District Court’s finding that Nexus Pharma’s generic, diluent- and magnesium-containing antibiotic composition would infringe.
The second basis for Nexus Pharma’s challenge to the District Court’s decision was that the specification did not satisfy the disclosure requirements under 35 U.S.C. § 112(a) because the relevant claim limitation (“injection site hemolysis of red blood cells is reduced relative to intravenous administration of a composition that does not include magnesium”) did not make clear that the comparison would be made against compositions that contained no divalent cations or ones containing a different divalent cation (such as calcium, for example). Nexus contended that to be consistent with satisfaction of the written description requirement the comparison must be between compositions having magnesium and that do not. The inventors failed to show that they had possession of reduced hemolysis when the comparison was made with compositions having calcium cations.
Melinta first argued that Nexus had waived this argument because it had appeared in only a single sentence of Nexus Pharma’s post-trial brief (which the Court rejected based on, inter alia, expert trial testimony by the parties on this issue). Melinta also argued that the correct comparison would be between compositions with magnesium and those without. Regarding this argument the Federal Circuit held that the District Court did not err and considered it to be the most natural comparison” between the same composition with and without magnesium (wherein the latter was Melinta’s prior formulation). The specification sets forth comparisons between formulations having divalent cations (including those with magnesium or calcium) and those without (i.e., lacking any metal cations). However, “[n]othing in the specification is directed to comparing the effect of magnesium with that of calcium [but r]ather, the focus of the specification is the reduction of hemolysis resulting from the inclusion of magnesium or calcium, as compared to prior art formulations, which did not include metal cations at all.”
The panel conceded that the District Court’s construction “does not expressly exclude a formulation that lacks magnesium but includes a different cation, such as calcium, in place of the absent magnesium.” Nevertheless, the panel opined that the construction argued by Nexus Pharma “would be contrary to the overwhelming evidence that the invention of the ‘802 patent was directed to the advantage of adding magnesium to a minocycline formulation lacking divalent cations, and not to any advantage magnesium might have compared to other metal cations.” It did not escape the panel’s notice that the prior art Minocin product included a warning against administering formulation including a calcium-containing diluent which may form a precipitate.
Having affirmed the District Court’s decisions on infringement and non-invalidity, the opinion explains that the parties had stipulated that the district court’s infringement finding would be upheld if either the ‘802 or ‘105 patents were found not invalid and thus the Court did not address validity of the ‘105 patent.
The outcome in this case is somewhat in contrast to that in Enanta Pharmaceuticals inc. v. Pfizer Inc. In that case, the patentee lost priority to its earliest-filed provisional application (U.S. Patent Application No. 63/054,048) for reciting alternatives for a substituent as being “NHC(O)—C2-C12-alkyl” whereas U.S. Patent No. 11,358,953 at issue before the Court recited “NHC(O)—C1-C12-alkyl.” The District Court’s decision was affirmed by the Federal Circuit, but it is just as easy to see that the decision could also have readily rested on a finding that the lack of disclosure in the ‘048 provisional precluded Enanta from a finding that Pfizer infringed by including a species (nirmatrelvir) in its product PaxlovidÒ) that had a C1 substituent (-CH3), and permitting Enanta to retain the scope of the C2-C8 alkyl species disclosed in the provisional application to which they were arguably entitled. Here the District Court and the Federal Circuit somewhat overlooked Nexus Pharma’s distinction that although the specification disclosed the need to add a diluent to the magnesium-containing composition prior to administration that wasn’t what was recited in the asserted claims of either patent-in-suit. Maybe that is because the asserted claims were method claims and the “invention” was inclusion of magnesium to a well-known antibiotic that reduced an important drawback, injection site hemolysis of red blood cells, rather than the composition of matter claims in Enanta. But it seems prudent to recognize the distinctions.
The U.S. Patent and Trademark Office has published a final rule in the Federal Register (an unpublished version of the notice, which will be published on June 24, can be found here), in which the Office states that it is “revising its practice of requiring additional information for delays in taking certain actions in patent applications and patents from requiring additional information for delays exceeding two years to requiring additional information for delays exceeding one year.”
The notice indicates that in a prior Federal Register notice published a little more than six years ago, the Office had clarified its practice with respect to situations requiring additional information about whether a delay in seeking the revival of an abandoned application, acceptance of a delayed maintenance fee payment, or acceptance of a delayed priority or benefit claim was unintentional (see “USPTO Issues Clarification Regarding Petitions Based on Unintentional Delay“). In particular, the prior notice indicated that the Office would require additional information when a petition to revive an abandoned application was filed more than two years after the date the application became abandoned, or when a petition to accept a delayed maintenance fee payment was filed more than two years after the date the patent expired for nonpayment, or when a petition to accept a delayed priority or benefit claim was filed more than two years after the date the priority or benefit claim was due.
In the final rule published today, the Office notes that it “has decided to shorten the two-year period for requiring additional information to one year,” explaining that “[i]f the period of delay is more than one year, there is a sufficient concern that the entire delay may not be unintentional and thus warrant a requirement for an additional explanation of the circumstances surrounding the delay.” The Office also notes that the change from a two-year period to a one-year period “aligns with USPTO efforts to reduce application pendency and promote efficient patent operations by encouraging applicants to regularly monitor patent files and promptly take corrective actions when needed.” The notice states that “[p]rompt, corrective actions are more likely to be associated with unintentional delay and assist in streamlining examination.”
The notice cautions stakeholders that “[n]othing in this notice should be construed as an indication that the USPTO will only require additional information in consideration of a petition” filed more than one year after the date an application became abandoned, the date a patent expired, the due date of a priority or benefit claim, or the expiration of the time limit in connection with the requirements pertaining to an international design application.
The new rule will take effect on August 13, 2026, and will apply to any new petition filed after the effective date. The notice indicates that the two-year period triggering the fees under 37 C.F.R. §§ 1.17(m)(1) and 1.17(m)(2) will be changed to a one-year period.
In a nonprecedential opinion, the Federal Circuit reviewed summary judgment granted to accused infringer Abiomed that claims asserted by Maquet Cardiovascular LLC were not infringed, in Abiomed Inc. v. Marquet Cardiovascular LLC.
The case arose over claims to intravascular blood pump systems and methods for using them to provide heart support. U.S. Patent Nos. 7,022,100; 8,888,728; 9,327,068; 9,545,468; 9,561,314; and 9,597,437 were asserted; by the time the case came to trial, only claims 16 and 17 of the ‘100 patent were at issue. Claim 1 of the ‘468 patent was also set forth in the opinion to further illustrate relevant claim terms construed by the District Court, wherein said construction was reversed at trial and upheld on appeal:
1. An intravascular blood pump system, comprising:
an intravascular blood pump adapted to be guided to a predetermined location within the circulatory system of a patient by a guide wire and configured to provide leftheart support, the intravascular blood pump comprising a rotor having a rotor hub tapering in the distal direction, at least one blade extending radially outward from the rotor hub,
a catheter coupled to a proximal end of the intravascular blood pump, a purge lumen extending through the catheter and operatively arranged to deliver purge fluid towards the intravascular blood pump;
a cannula coupled to a distal end of the intravascular blood pump, one or more first ports and one or more second ports establishing fluid communication between a lumen of therefrom and located distal to the at least one first port, the cannula is configured such that when the intravascular blood pump is positioned in the patient to provide leftheart support the distal end of the cannula and the at least one second port are positioned inside the patient’s heart and the proximal end of the cannula and the at least one first port are positioned in the patient’s aorta, the intravascular blood pump is configured to draw blood from the patient’s heart into the at least one second port through the cannula lumen and out the at least one first port to provide leftheart support while the cannula is positioned across an aortic valve of the patient;
an elongate lumen associated with the cannula and sized to slidably receive the guide wire and dimensioned such that the guide wire passes slidably and coaxially through the elongate lumen, the elongate lumen is sized smaller cross sectionally than the cannula lumen, both the elongate lumen and the cannula lumen not extending through the rotor hub, the intravascular blood pump system configured for the guide wire to extend proximally away from the intravascular blood pump, the guide wire not passing through the rotor hub or the catheter, and the guide wire extending out of the intravascular blood pump system in a distal direction through the elongate lumen;
a pressure sensing element configured to sense pressure proximate the intravascular blood pump;
a housing connected to a proximal end of the catheter; and first and second conduits each connected to the housing, at least one of the first conduit and second conduit in fluid communication with the purge lumen.
16. An intravascular blood pump system comprising:
an intravascular blood pump having a cannula coupled thereto,
a guide mechanism adapted to guide said intravascular blood pump and cannula to a predetermined location within the circulatory system of a patient, and
a blood pressure detection mechanism to detect the pressure of the blood proximate
at least one of the intravascular blood pump and cannula.
17. The intravascular blood pump system of claim 16 and further, wherein said blood pressure detection mechanism comprises at least one of fluid filled column disposed within at least a portion of said cannula, a piezoelectric element coupled to at least one of the intravascular blood pump and cannula, and a strain gauge coupled to at least one of the intravascular blood pump and cannula.
(Wherein italicized text identifies the elements of the patent claims in dispute.)
Abiomed initially filed a declaratory judgment action that its intravascular blood pump systems did not infringe any of the claims of the asserted patents and that those claims were invalid; Maquet filed a counterclaim asserting infringement and demanding money damages.
The opinion recognizes that the asserted claims disclosed three “broad aspect[s] of the present invention,” each defined by a specific type of “guide mechanism:” (1) “an ‘over-the-wire’ type guide mechanism;” (2) “a ‘side-rigger’ or ‘rapid exchange’ type guide mechanism;” and (3) “a ‘guide catheter’ type guide mechanism.” These embodiments were recited in claims with three claim limitations at issue: (1) “an elongate lumen associated with the cannula;” (2) “purge fluid;” and (3) “guide mechanism,” with the “guide mechanism” claim being recited in the asserted claims of the ‘100 patent.
The District Court issued a claim construction order wherein the limitation “an elongate lumen associated with the cannula” in the ‘468, ‘314, and ‘437 patents was construed to mean “a permanent elongate lumen formed along the side of the cannula.” The term “purge fluid” in the ‘728, ‘068, ‘468, ‘314, and ‘437 patents was construed to require that “the purge fluid does not go through the rotor bearings and into the bloodstream.” Finally, the District Court construed the limitation “guide mechanism” in the ‘100 patent to be in means-plus-function format. Maquet limited the asserted claims to claims 16 and 17 of the ‘100 patent as set forth above as a result of the District Court’s claim construction.
The District Court held on summary judgment that Abiomed’s accused infringing intravascular blood pump systems did not infringe asserted claims 16 and 17. This appeal followed.
The Federal Circuit affirmed-in-part, vacated-in-part, and remanded, rejecting the four arguments asserted by Maquet. The first of these was that the District Court erred in construing the phrase “elongated lumen associated with the cannula” to be limited to those elongated lumens that are “formed alongside of the cannula.” Maquet did not argue that there was error in the District Court’s determination that nothing in the plain language of the claims, the specification, or the prosecution history supported restricting the term to lumens found alongside the cannula. Rather, the District Court’s relied upon statements made by Maquet in inter partes proceedings, which Maquet argued was error. The Federal Circuit panel agreed, citing Phillips v. AWH Corp., 415 F.3d 1303, 1313–14 (Fed. Cir. 2005) (en banc), for the canons of construction that claims are construed under their plain and ordinary meaning to the skilled artisan, the specification and the prosecution history. There are two exceptions recognized by the Court: the first is when a patentee acts as their own lexicographer and provides express definitions; and the second is when a patentee disclaims the full scope of a claim in the specification or during prosecution. The panel assessed each of the three Philips bases of claim construction and found no support for the District Court’s construction of this phrase in the claims. For example, while there are species embodiments of this construction in the specification these are “exemplary only” and don’t limit the scope as construed by the District Court.
The inter partes review source of the District Court’s claim construction of this term was based on Marquet’s statements distinguishing its claims from a reference to Jegaden in those proceedings. The District Court’s error, in the panel’s view, was interpreting these statements as constituting prosecution disclaimer. Such an interpretation is not forbidden under Aylus Networks, Inc. v. Apple Inc., 856 F.3d 1353, 1360 (Fed. Cir. 2017), but in order for prosecution disclaimer to be properly invoked under these circumstances those statements must satisfy both of two requirements, specifically that the disclaimer be clear and unmistakable under Omega Eng’g, Inc. v. Raytek Corp., 334 F.3d 1314, 1326 (Fed. Cir. 2003). The Federal Circuit did not consider Maquet’s statements to have been “words or expressions of manifest exclusion or restriction” under Unwired Planet, LLC v. Apple Inc., 829 F.3d 1353, 1358 (Fed. Cir. 2016) (quoting Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1327 (Fed. Cir. 2002)), and were “far too slender . . . reed[s] to support the judicial narrowing of a clear claim term,” citing N. Telecom Ltd. v. Samsung Elecs. Co., 215 F.3d 1281, 1294 (Fed. Cir. 2000). Thus, the panel held the District Court’s construction of this term was error.
Maquet’s second argument was that the District Court erroneously imported a negative limitation into the term “purge fluid,” specifically that the fluid “does not go through the rotor bearings and into the bloodstream.” The panel rejected Abiomed’s support for this construction based on distinctions made to overcome the prior art (International Publication No. WO 99/02204) and disparagement thereof, and held that the District Court erred. As in its analysis of the cannula term, the opinion parsed the Philips factors and found wanting the District Court’s construction using this assessment. In fact, the panel found evidence from the specification that permitted the purge fluid to pass though the components of the claimed intravascular blood pump system and into the bloodstream. The District Court, while recognizing these deficiencies nevertheless found that Maquet had “unmistakably disparaged” configurations permitting the purge fluid to enter the bloodstream, again arising from statements made during an IPR. Once again, the panel held that the statements did not “amount to an instance ‘[w]here an applicant argues that a claim possesses a feature that the prior art does not possess in order to overcome a prior art rejection,’” citing Seachange Int’l, Inc. v. C-COR, Inc., 413 F.3d 1361, 1372–73 (Fed. Cir. 2005), and also Andersen Corp. v. Fiber Composites, LLC, 474 F.3d 1361, 1374 (Fed. Cir. 2007). Under these circumstances, where in addition it was unclear on the record whether the argument was made to distinguish the prior art, the panel held the District Court erred in finding “clear and unmistakable” disclaimer and vacated this determination, remanding to the District Court “for proceedings consistent with [the] opinion.”
Third, Maquet argued that the District Court erred in construing the term “guide mechanism” as being a means-plus-function term. On this construction the panel agreed with Abiomed that the District Court had not erred. The Federal Circuit recognized that the term did not contain the canonical “means for” terminology but also that under Williamson v. Citrix Online, LLC, 792 F.3d 1339, 1346 (Fed. Cir. 2015), that language was not mandatory. Under Williamson, the presence or absence of this phraseology raised a rebuttable presumption for or against (respectively) a finding that 35 U.S.C. § 112 ¶ 6 was to be applied in construing the claim. In this case, the panel held the District Court had not erred because the claim did not recite sufficiently definite structure and the skilled worker would not have understood the claimed element (the guide mechanism) to possess “a sufficiently definite meaning as the name for a structure” under Williamson. In the Court’s assessment, “mechanism” is nothing other than a “nonce” word and “guide” is a merely a “functional modifier,” citing Media Rts. Techs., Inc. v. Cap. One Fin. Corp., 800 F.3d 1366, 1373 (Fed. Cir. 2015). The remaining claim language does not remedy these deficiencies according to the opinion as the claim only recited functional terms. Nor did the Court find any “clear and unambiguous” definition for the term in the specification.
Having decided that the District Court had not erred in considering “guide mechanism” to be in means-plus-function format, the opinion determined that the specification was lacking in the required “sufficient structure” that “corresponds to the claimed function.” The District Court held that the structural disclosures in the specification amounted to:
(a) a guide wire passing slideably through a central lumen extending through a drive cable assembly, blood pump, and cannula; (b) a guide wire passing slideably through a lumen extending through a guide carriage integrally formed along at least a portion of the cannula sidewall; or (c) a conduit assembly, including guide catheter, a rotor shroud, and a cannula, which is capable of docking to a separate pump assembly.
Maquet objected to this description by arguing that the District Court failed to include in this litany “(1) a ‘rapid exchange’ guide mechanism and (2) a guide mechanism that ‘includes a guide carriage 124.’” The Federal Circuit rejected these objections because Maquet supported its argument with extrinsic evidence which the panel found insufficient under Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1584 (Fed. Cir. 1996).
Finally, Maquet argued that granting summary judgment in Abiomed’s favor was error because there remained genuine issues of material fact even if the Federal Circuit held that the District Court had properly construed the term “guide mechanism.” Once again, here the Federal Circuit agreed with Abiomed, reciting the rubric that “[t]o literally infringe a means-plus-function limitation, ‘the accused structure must either be the same as the disclosed structure or be a section 112, paragraph 6 “equivalent,” i.e., (1) perform the identical function and (2) be otherwise insubstantially different with respect to structure,’” citing Kemco Sales, Inc. v. Control Papers Co., 208 F.3d 1352, 1364 (Fed. Cir. 2000). The District Court had properly considered summary judgment by viewing the evidence in light most favorable to non-movant Maquet and accepting Maquet’s specific factual contentions. The basis for this reasoning was illustrated by drawings showing the arrangement of the disputed elements in the claimed device:
and stating “[s]pecifically, the parties dispute whether the structure of the Impella products meets structure (b) of the district court’s claim construction. Maquet contends that ‘a jury could reasonably find that the Impella’s pigtail guide carriage is “integrally formed along at least a portion of the cannula sidewall.’” However, as the District Court determined, “the pigtail is not attached to any portion of the cannula sidewalls. Rather, it is attached to the threaded nub, which is attached to a flat surface, which is attached to the inlet cage at the distal end of the cannula[:]”
Because “[t]he cannula sidewalls lie entirely below (proximal) to that plane; the pigtail and threaded nub lie entirely above (distal) to it, . . . there is never any point in the structure where the pigtail or nub are ‘along’ even a ‘portion’ of the cannula sidewall.”
The opinion thus asserts that accordingly “the district court did not reversibly err in determining that the Impella pigtail is not ‘formed along at least a portion of the cannula sidewall.’”
Finally, in this regard, the opinion rejects Maquet’s assertion of expert testimony based on its characterization of that testimony as being “conclusory” and “insufficient to defeat summary judgment,” citing Traxcell Techs., LLC v. Sprint Commc’ns Co., 15 F.4th 1121, 1129 (Fed. Cir. 2021).
In sum, the Federal Circuit affirmed-in-part, vacated-in-part, and remanded to the District Court.
In a nonprecedential decision, the Federal Circuit recently affirmed a district court’s finding that innovator drugmaker Otsuka Pharmaceutical failed to show generic competitor Lupin infringed (or would infringe, if it marketed an FDA-approved generic version of Otsuka’s JYNARQUE® drug), the claims of U.S. Patent No. 8,501,730 and that Lupin had shown that Otsuka’s method of making the active pharmaceutical agent, highly pure tolvaptan, was obvious in Otsuka Pharmaceutical Co. v. Lupin Ltd.
In ANDA litigation, Otsuka asserted claims of U.S. Patent Nos. 8,501,730 and 8,273,735:
U.S. Patent No. 8,501,730:
1. A highly pure 7-chloro-5-hydroxy-1-[2-methyl-4-(2-methylbenzoylamino)benzoyl]-2,3,4,5-tetrahydro-1H-1-benzazepine having a purity of more than 99.5%, or a salt thereof, which is produced by the process which comprises reducing a benzazepine compound of the formula (1):
wherein X1 is a halogen atom, R1 and R2 are independently a lower alkyl group, or a salt thereof in the presence of a hydrogenating agent selected from the group consisting of lithium aluminum hydride, sodium borohydride, zinc borohydride, and diborane in an amount of 0.25 to 1 mole per 1 mole of the compound (1).
U.S. Patent No. 8,273,735:
6. A process for producing a 2,3,4,5-tetrahydro-1H-1-benzazepine compound of the formula(10):
wherein X1 is a halogen atom, R1 and R2 are independently a lower alkyl group, or a salt thereof, which comprises reducing a benzazepine compound of the formula (1):
wherein R1, R2 and X1 are as defined above, or a salt thereof in the presence of a hydrogenating agent selected from the group consisting of lithium aluminum hydride, sodium borohydride, zinc borohydride, and diborane in an amount of 0.25 to 1 mole per 1 mole of the compound (1).
7. The process according to claim 6, wherein the hydrogenating agent is sodium borohydride which is used in an amount of 0.25 to 1 mole per 1 mole of the compound (1).
(wherein dependent claims 7, 8, and 10 were asserted).
Otsuka’s drug product and method for producing it had the benefit of reducing the percentage (to be less than 0.05%) of an impurity (termed “the dechlorinated impurity” in the opinion) for the tolvaptan used to treat Autosomal Dominant Polycystic Kidney Disease (ADPKD). The District Court found that Lupin’s Drug Master File differed from Otsuka’s method because “where Otsuka’s claimed process generally uses 1 molar equivalent or less of hydrogenating agent such as sodium borohydride per 1 mole of precursor compound, Lupin’s process uses at least 1.2 molar equivalents of sodium borohydride per 1 mole of precursor” (emphasis in opinion).
In other differences in its commercial process, Lupin tested the progress of the synthesis reaction at 15 and 75 minutes after initiation to determine whether no more than 0.05% of the original amount of the precursor compound remained. If that was the case, the reaction was quenched by the addition of hydrochloric acid and water, whereas if 0.05% of the original amount of the precursor compound or more remained additional reducing agent was added. On the basis of these differences, the District Court held Lupin’s DMF method does not infringe Otsuka’s claimed (and approved) method.
Regarding invalidity, the District Court found that the asserted claims of the ‘735 patent were obvious over the cited prior art while the asserted claims of the ‘730 patent were not.
The Federal Circuit affirmed, in an opinion by Circuit Judge Hughes, joined by Circuit Judge Cunningham and the Honorable Allison D. Burroughs, District Judge, U.S. District Court for the District of Massachusetts, sitting by designation.
In its assessment of the District Court’s infringement determination, the panel reviewed the District Court’s conclusion for clear error regarding fact findings and its legal conclusions de novo, citing United Therapeutics Corp. v. Liquidia Techs., Inc., 74 F.4th 1360, 1367 (Fed. Cir. 2023), and Merck Sharp& Dohme Corp. v. Amneal Pharms. LLC, 881 F.3d 1376, 1384 (Fed. Cir. 2018). Otsuka argued that the District Court’s reasoning was disjointed with regard to the claim term “amount,” wherein the District Court adopted a “practical completion” for construction and a “different and contradictory” construction (“absolute completion”) in applying the terms to Lupin’s DMF. Otsuka contended that “the reaction is practically complete before 1 molar equivalent of hydrogenating agent is added to the reaction mixture.” The District Court had construed the term “amount” to mean “the amount [of hydrogenating agent] while the reaction is taking place.” The panel also considered the differing definitions asserted by the parties during post-trial briefing, where Otsuka argued that the amount of hydrogenating agent recited in the claims was “the amount of hydrogenating agent added before the reduction reaction reaches ‘practical completion’ or is ‘complete in a practical sense,’” and Lupin’s position was that “so long as unreacted ketone precursor and sodium borohydride are present, the reaction will continue to proceed.” Otsuka’s argument prevailed with the District Court, on the grounds that Lupin’s position was directed to “the ‘theoretical possibility’ of ‘absolute completion’ of the reaction.”
On the basis of this construction, the District Court found that Otsuka’s infringement contentions failed for two reasons. First, Otsuka’s evidence did not establish that the reduction reaction was “practically complete” by the time the amount of unreduced precursor fell to 0.05% of the original amount in the reaction mixture when Lupin’s recipe ended the reaction by initiating the quenching step. Second, should Lupin’s reaction reach that amount Otsuka failed to show the other limitations of the claims were satisfied, that “no more than 1 molar equivalent of sodium borohydride or less has been added by that point.” The panel, in affirming these determinations, rejected Otsuka’s evidence to the contrary, which consisted of the results of two experiments performed by Lupin during the development of their synthetic process, and testimony from its own expert that “the reduction reaction will be complete long before the addition of 1 molar equivalent of sodium borohydride.” The basis for the Court’s rejection of these arguments was that, for Otsuka’s first contention the data from the two experiments contained “anomalies” that could be accounted for by “some unspecified ‘margin of error’ or a lack of ‘appropriate quality control.’” These anomalies were enough for the panel to agree with the District Court that reliance on these experiments by Otsuka’s expert was faulty because the data itself was “flawed.”
Otsuka’s second argument, based on its expert’s testimony, “that the reduction reaction will be complete long before the addition of 1 molar equivalent of sodium borohydride” was also unpersuasive to the Federal Circuit. Because some of these conclusions relied on the flawed experiments from Otsuka’s first arguments the panel agreed with the District Court that that Lupin’s expert’s testimony was more reliable and consistent with the evidence than Otsuka’s evidence.
Another Otsuka argument that did not convince the panel was that the District Court had erred with regard to Lupin’s expert witness testimony and purported deficiencies thereof. Specifically, Otsuka argued that Lupin’s expert was not properly qualified as an expert for failing to satisfy the standard of having “ordinary skill in the art,” based on lack of experience incorporated into the definition adopted by the District Court (having a relevant doctorate degree and ‘at least two years of experience in the synthesis, research, and development of medicinal compounds’”). The opinion notes that Otsuka had made these same arguments during post-trial briefing but the District Court refused to “entertain the merits” for failure to object and thus preserve challenge to admissibility of Lupin’s expert’s testimony on this basis. The panel found that the District Court properly determined that Otsuka had forfeited this objection under the law of the Third Circuit under Harbor Bus. Compliance Corp. v. Firstbase.io, Inc., 152 F.4th 516, 527 (3d Cir. 2025); United States v. Rivas, 493 F.3d 131, 136 (3d Cir. 2007). In addition, the panel cited Otsuka’s failure to file a Daubert motion under these circumstances, Daubert v. Merrell Dow Pharmaceuticals (1993). On these bases, the Federal Circuit affirmed the District Court’s conclusion that Otsuka had failed to establish infringement.
Regarding the District Court’s invalidity holding, the opinion sets forth the bases on the District Court record for that determination. These include a reference to Kondo et al.* that teaches small-scale synthesis of tolvaptan, which the District Court found would provide motivation for the skilled artisan “to specifically select the reduction step for modification” based on testimony from Lupin’s expert. Having so chosen, that artisan would have been motivated to “reduce the amount of sodium borohydride used in Kondo’s reduction step” and to achieve the claimed invention either “in light of his or her background knowledge or in light of two other [unidentified herein] pieces of prior art.” The District Court also found that these motivated alterations in prior art synthetic methods would have been pursued by the skilled artisan with a reasonable expectation of success.
The Federal Circuit, in affirming these conclusions by the District Court, rejected Otsuka’s arguments based on Lupin’s expert witness’s deficiencies in his focus on the Kondo reference and its reduction step as an indication of hindsight bias, as well as dismissing secondary considerations including “unexpectedly high tolvaptan purity and yield” over the Kondo prior art. The panel found that Lupin’s expert in his testimony considered other prior art but found Kondo to be “a particularly promising reference.” Also significant for the panel was that a non-prior art reference referred to the Kondo reference to be “a standard method for synthesizing tolvaptan.” The Federal Circuit concurred, despite acknowledging that there were “major obstacles” in developing the method. The Federal Circuit further found no clear error in the District Court’s assessment of the motivation to combine the prior art. This conclusion appears to be due in part to both parties’ experts having identified “many of the steps of Kondo” to have been “considered for modification,” including the reduction step. This conclusion was bolstered by Lupin’s “significant” expert testimony as to why the skilled worker would have done so, including “cost savings, improvement of the reaction’s safety profile, post-processing simplification, and knowledge that the hydrogenating agent was already being used in molar excess.”
As for the secondary considerations aspect of the District Court’s obviousness determination, the panel noted that Otsuka’s expert testimony was not based on Kondo’s methods but on modifications that used double the amount of reducing agent (emphasis in opinion). The opinion also asserts that the purported evidence of “unexpectedly high [product] yield” was inaccurate (comparing 30% to the 82-93% yield using the claimed process). In any event, the Federal Circuit dismissed secondary considerations errors by the District Court (should they exist) as being harmless, on the basis (“[a]fter all”) that “weak secondary considerations generally do not overcome a strong prima facie case of obviousness,” citing Genentech, Inc. v. Sandoz Inc., 55 F.4th 1368, 1378 (Fed. Cir. 2022).
* Kazumi Kondo et al., 7-Chloro-5-hydroxy-1-[2-methyl-4-(2 methylbenzoyl amino)benzoyl]-2,3,4,5-tetrahydro-1H-1-benzazepine (OPC-41061): A Potent, Orally Active Nonpeptide Arginine Vasopressin V2 Receptor Antagonist, 7 Bioorganic & Med. Chemistry 1743, 1743–54 (1999).
The Supreme Court issued an Order this morning denying certiorari in Newman v. Moore. The Court’s Order states that:
The motion for leave to file a brief in opposition under seal with redacted copies for the public record is granted. The motion for leave to file a reply under seal with redacted copies for the public record is granted. The petition for a writ of certiorari is denied.
Patent Docs 