GlaxoSmithKline LLC v. Teva Pharmaceuticals (Fed. Cir. 2020)

GSK’s patent at issue here covers a method of treating congestive heart failure with the drug carvedilol (Coreg) along with an ACE inhibitor, a diuretic  and digoxin.  RE40,000. Claim 1 below is representative and shows the limitation added during the reissue process:

1. A method of decreasing mortality caused by congestive heart failure in a patient in need thereof which comprises administering a therapeutically acceptable amount of carvedilol in conjunction with one or more other therapeutic agents, said agents being selected from the group consisting of an angiotensin converting enzyme inhibitor (ACE), a diuretic, and digoxin,

wherein the administering comprises administering to said patient daily maintenance dosages for a maintenance period to decrease a risk of mortality caused by congestive heart failure, and said maintenance period is greater than six months.

RE40,000 (original US Pat 5,760,069). Each of these drugs were already known for treating heart disease, and the narrowing reissue was filed after Teva first challenged the patent.

Teva did a couple of things to avoid infringement. First, Teva waited until 2007 to launch its product — that was when the underlying patent on the drug carvedilol expired.  Second, Teva attempted to avoid directly market its drug for the purposes of treating congestive heart failure.  In particular, its product labelling focused instead on hypertension and Ventricular Dysfunction following MI (two non-patented approved uses of the drug).   However, in 2011 the FDA required Teva to alter its labelling to be identical-in-content to the approved GSK product.  Thus, in 2011 Teva added treatment for heart failure as an indication for treatment.

Once the labeling was changed, GSK then sued for inducing doctors to infringe.

271(b) Whoever actively induces infringement of a patent shall be liable as an infringer.

At trial the jury sided with the patentee — finding the patent valid and willfully infringed and then awarding $234 million in lost profit damages.

Post-trial, the district court flipped the award — finding insufficient evidence that Teva “actually caused” any particular physician to infringe. “Without proof of causation, which is an essential element of GSK’s action, a finding of inducement cannot stand.”  District Ct. Awarding JMOL.  The district court noted lots of publications and promotions informing physicians about how to use the product to treat heart failure — none of which came from Teva.  Practicing doctors don’t read the fine-print on the side of the bottle. In fact, GSK’s MD expert admitted that he did not read the Teva label before prescribing.

On appeal here, the Federal Circuit has reversed — holding that the circumstantial evidence of inducement was sufficient.  In particular, in this situation Doctors relied upon Teva’s statements that its drug was interchangeable with GSK’s — a “complete replacement.”  My Mizzou colleague, Professor Erika Lietzan testified as a FDA-expert for GSK at trial. Her testimony includes the conclusion that (1) the “AB-rating” of the generic indicates that the products are interchangeable; and (2) comparing GSK’s product by-name creates an implication of interchangeability.

The majority opinion here was written by Judge Newman and joined by Judge Moore. Chief Judge Prost wrote in dissent — arguing that the decision here undermines the balance between innovator incentives and the introduction of lower-cost generics.

Teva waited until GSK’s patent covering the carvedilol compound expired to launch its product covering two unpatented indications hypertension and post-MI LVD. So, when GSK’s ’000 reissue patent later issued—reciting a narrow method of treating a third indication, CHF—Teva’s skinny label did not even suggest using its product according to the patented method.

At the FDA’s direction, Teva amended its label years later to include the patented method, but there was still no inducement via the full label. Nothing changed in the market, and doctors’ prescribing decisions were not affected. By that time, GSK could not rely on Teva’s ANDA as an artificial act of infringement. Thus, to prove induced infringement, GSK had to show that Teva actually caused doctors to directly infringe the ’000 patent. It failed to do so.

Dissent.  With regard to Prof. Lietzan’s testimony, the dissent quoted to her same statement regarding AB-rating — noting that “AB rating reports therapeutic equivalence only ‘if the generic drug is used in accordance with the label.'” The AB-rating does not apply to “off-label uses.”

= = = =

The majority also sustained the jury’s damages verdict.  Teva argued that the lost-profit award was improper because there were other generic versions on the market.  If Teva hadn’t been on the market then the sales would have gone to the other generic producer — not back to the higher-priced GSK.  In its decision, the district court identified the other generic versions as “infringing alternatives” (GSK is in a separate lawsuit against them) and thus should not count in the economic picture of lost profits.  On appeal, the Federal Circuit affirmed — holding that “The district court correctly instructed the jury that the availability of carvedilol from other generic producers is not a ‘noninfringing substitute.'”

Thus, in the end, the jury verdict is reinstated and Teva will have to pay the money. Note — the ‘000 patent is also expired and so generics continue to be available.

= = = =

Presumably, Teva could have avoided infringement by sending notices to pharmacies and doctors to avoid proscribing its product for congestive heart failure.

The following question is one I taught in Civil Procedure earlier this semester:

May a motion for Rule 11 sanctions be granted when the moving party fails to serve the motion on the nonmoving party prior to filing?

In my class, I taught that R.11(c)(2) prohibits a party from filing a R.11 motion for sanctions without first serving “the motion” on the non-moving party (and waiting 21 days).

• My students: What if we the moving party provides notice by sending an email to say that they are planning to file a motion and explaining the reasons?
• Crouch: No, the rule says that you have to serve “the motion;” not just provide notice of your concern.

In Khan v. Hemosphere, Inc., the Federal Circuit disagreed with my analysis of the rule. In particular, the court held that the Khans were sufficiently “on notice of [defendants’] intent to seek sanctions” based upon a series of letters sent to the Khans indicating that a sanctions-motion was coming. Thus, although the Khans were not served with “the motion” they were sufficiently on notice.

Khan Petition for Rehearing: In their petition, the Khans argue that the Federal Circuit’s “conclusion that warning letters of the type at issue here can take the place of the ‘motion’ required by Rule 11(c)(2) breaks sharply with the text of the Rule, and with every other Court of Appeals to consider the issue—including the Seventh Circuit.” Here, the 7th Circuit is important because this case arose in Illinois and the regional circuit’s law should apply to this non-patent related issue.  The 7th Circuit is also important because it is an outliner in allowing “substantial compliance” with the R. 11 rather than sticking to “the motion” rule. The 7th Circuit’s approach appears to have begin with a flippant ruling by Judge Easterbrook in Nisenbaum v. Milwaukee County, 333 F.3d 804, 808 (7th Cir. 2003). In that case the court decided without any analysis that “substantial compliance” with the notice requirement is sufficient.

In its decision, the Federal Circuit cited to Matrix IV, Inc. v. Am. Nat’l Bank & Tr. Co., 649 F.3d 539, 552–53 (7th Cir. 2011), which stated that “a letter informing the opposing party of the intent to seek sanctions and the basis for the imposition of sanctions” was sufficient.  In its petition for rehearing, Khan explains that the Matrix IV statement was dicta and insufficient to overcome other 7th Circuit decisions which require the to-be-sanctioned-party an “opportunity to withdraw or correct the challenged pleading within 21 days without imposition of sanctions.”  Khan argues that the 21-day opportunity was not given here.

Wright & Miller suggest that informal notice as provided here is insufficient:

Note that informal notice—rather than formal service—of a potential violation is insufficient to trigger the beginning of the twenty-one day safe harbor period. The Advisory Committee Note explains that although informal notice does not trigger the safe harbor period, it usually is expected that informal notice will be given before a party prepares and serves a formal motion under Rule 11 for sanctions.

§ 1337.2Procedural Aspects of Rule 11 Motions—The Safe Harbor Provision, 5A Fed. Prac. & Proc. Civ. § 1337.2 (4th ed.).

I hope the Khans win here, otherwise I’ll be forced to change my syllabus.

by Dennis Crouch

Biogen MA, Inc. v. EMD Serono, Inc. & Pfizer Inc. (Fed. Cir. 2020) [BiogenSerono]

After a five-week-trial, the jury returned a verdict that Biogen’s asserted claims were anticipated by two prior art references. In the lawsuit, Biogen had asserted infringement of its US7588755 against Serono and Pfizer based upon their sale of Rebif (IFN-β used for MS treatment).

In a post-judgment order, the District Court rejected this portion of the jury verdict–holding that no reasonable jury could have found anticipation. In addition to JMOL, the district court also conditionally granted a new trial on anticipation under R.59.  The jury had sided with Biogen on other grounds of infringement/validity and so it looked like a win for the patentee. Because the original jury had found the patent invalid, it did not award any damages.  Thus, the district court entered a “partial judgment” and scheduled a new trial on damages.

Appeal before a Damages Trial: 28 U.S.C. § 1292(c)(2), provides appellate jurisdiction once a patent case is “final except for an accounting.”  In Robert Bosch, LLC v. Pylon Mfg. Corp., 719 F.3d 1305 (Fed. Cir. 2013)(en banc) the court determined that “accounting” as used in that provision included a jury trial on damages.  Thus, appeal was appropriate at this point.

Product-by-Process Within a Method: On appeal, the Federal Circuit agreed with the jury that prior administration of native IFN-β anticipates the claims here.  The district took issue with the conclusion since the claims expressly require treatment using “recombinant” inf-β produced in a “non-human host” and that had been “transformed by a recombinant DNA molecule.”

Although the claim is a method claim, it requires use of a particular product (IFN-β) produced by a particular process (recombinant transformation in a non-human host). Of note, the claim does not appear to require the manufacturing step, only that administration of a product created in that manner.  On appeal, the Federal Circuit determined that this setup is appropriate for the product-by-process rule: an old product is not patentable even if it is made by a new process. See Gen. Elec. Co. v. Wabash Appliance Corp., 304 U.S. 364 (1938).

In applying the product-by-process rule to a method claim, the court reasoned as follows:

If the novelty of the recombinant IFN-β composition requires comparing its structure to the structure of native IFN-β, as Amgen requires, it would defy all reason to excuse that analysis for a method of administration claim using that composition. Such a rule could have the absurd result that a recombinant composition could be non-novel, the method of administration could be non-novel, but the method of administration of the composition defined by the process of its manufacture would be novel as a matter of law.

There is no logical reason why the nesting of a productby-process limitation within a method of  treatment claim should change how novelty of that limitation is evaluated. Indeed, we have previously applied product-by-process analysis to a nested limitation. . . . The nesting of the product-by-process limitation within a method of treatment claim does not change the proper construction of the product-by-process limitation itself.

Slip Op. (Citing Purdue Pharma L.P. v. Epic Pharma, LLC, 811 F.3d 1345 (Fed. Cir. 2016)).

Biogen also argued that the prior art was inadequate because there was no proof that it was had the same structure.  Although the native molecule had the same polypeptide sequence, there was no evidence that it had folded in the same manner — and thus did not necessarily meet the “therapeutically effective amount” limitation.  On appeal, the court noted that the prior art was the same polypeptide sequence as required by the claim and had also shown therapeutic activity — the claim did not require that it be the identical therapeutic activity or identical folding pattern as the patentee’s product.

After reviewing these issues, the appellate panel found the jury had a reasonable basis for its invalidity decision.

= = =

The new trial issue is a bit trickier.  The rule states that the court my grant a new trial “for any reason for which a new trial has heretofore been granted.” R. 59.  Here, the new trial was apparently justified on “the same legal errors” as the JMOL determination. “None of the additional considerations noted by the district court in support of its conditional grant of a new trial are independently sufficient to support its decision.”  Thus, the new trial is also reversed.

On remand, the defendant wins and the patent claims are invalid.

by Dennis Crouch

Network-1 Techs. v. Hewlett-Packard Co. (Fed. Cir. 2020)

In this case, the E.D. Tex. jury came back with a win for the defendant — finding that HP did not infringe Network-1’s US6218930 and that the patent was invalid.  Post-Verdict, the district court flipped on validity — holding that HP was estopped from raising its obviousness challenge because it had joined an (unsuccessful) IPR against the patent.

The invention here relates to logic for sending a power-supply on the same twisted-pair used for data transmission (Power over Ethernet or PoE). The basic approach is that the ethernet cable will start-off with a low-level current. If a particular access-device signals that it can handle higher power, then the server will raise the power level being sent.  The patent here issued in 2001 (1999) priority and the lawsuit was originally filed in 2011.

Statutory Estoppel: The big issue in the case for the patent world is statutory estoppel. One reason why this lawsuit took so long to complete was that there were two intervening reexaminations and one inter-partes-review.  HP was time-barred from bringing its own IPR, but was able to join one filed by Avaya.  The instituted IPR challenged the patent claims for anticipation and obviousness based upon two prior art references Matsuno and De Nicolo. Although the PTAB granted the IPR, it eventually upheld the validity of the challenged claims.

One result of losing an IPR challenge is estoppel under 35 U.S.C. § 315(e)(2).

(2) Civil actions and other proceedings. The petitioner in an IPR … that results in a final written decision … may not assert … that the claim is invalid on any ground that the petitioner raised or reasonably could have raised during that inter partes review.

Id.  Following the IPR final written decision, the district court eventually determined that HP was generally estopped from raising an obviousness challenge (this ruling unfortunately came post-verdict after the obviousness challenge had already been raised).

On appeal, the Federal Circuit vacated — finding that the district court too broadly applied estoppel. In reading the statute, the appellate panel concluded that “a party is only estopped from challenging claims in the final written decision based on grounds that it ‘raised or reasonably could have raised’ during the IPR.”  Slip Op. In this case, HP was time-barred from brining new claims and was limited to simply joining the claims brought by Avaya.  “Because a joining party cannot bring with it grounds other than those already instituted, that party is not statutorily estopped from raising other invalidity grounds.” Id.

What is a “Ground.” The USPTO has a way that it treats a “ground” in inter partes review, but how should the term really be construed?  The estoppel provision applies to “any ground” raised in the IPR, but the statute does not go on to define the meaning of the term “ground.” Here, the court interpreted the term as any invalidity contention against the challenged claims in the IPR based upon the prior art that served as the basis for the IPR.  The court writes:

When the Board reached a final written decision …, HP was statutorily estopped from raising invalidity grounds based on Matsuno and De Nicolo against claims 6 and 9 in a district court action. HP, however, was not estopped from raising other invalidity challenges against those claims because, as a joining party, HP could not have raised with its joinder any additional invalidity challenges.

Slip Op. Remember, the jury found the patent invalid and that verdict was rejected by the court on JMOL (JNOV).  That JMOL decision has been vacated. On remand, the district court will now need to decide whether to (1) give effect to the verdict; or (2) hold a new trial on validity.

The Federal Circuit got this right according to the Statute — and it also serves as a signal to folks who are deciding whether to join a pending IPR that the estoppel consequences will be quite limited.

Claim Construction: On appeal, the Federal Circuit provides the patentee with potential shot at winning by shifting the claim claim construction. In particular, the court found error in construction of the term “main power source.”  In particular, the district court construed the phrase as requiring a “DC power source” and on appeal the Federal Circuit expanded the definition:

We conclude that the correct construction of “main power source” includes both AC and DC power sources. There is no dispute that the ordinary meaning of “power source” includes both AC and DC power sources. And neither the claims nor the specification of the ’930 patent require a departure from this ordinary meaning.

Slip Op.  On remand, we’ll see if a new jury changes its mind based upon this difference.

Claim Broadening: The third question on appeal involved HP’s argument that the patentee had improperly broadened claim 6 during a reexamination in a way prohibited by statute:

No proposed amended or new claim enlarging the scope of a claim of the patent will be permitted in a reexamination proceeding

35 U.S.C. § 305. A claim is “enlarg[ed]” if it covers any embodiments not covered by any original claim.

Some facts:

• Two prior district courts interpreted Claim 6’s “secondary power source” to be physically separate from the claimed “main power source.”
• In the reexamination, two dependent claims were added to to require that the two power sources be “the same physical device.”

Although claim 6 was not itself amended, the dependent claim strongly suggested that the claim should be interpreted differently. As the lawsuit progressed, Network-1 subsequently disclaimed these newly added claims and the district court gave Claim 6 its narrow interpretation.

On appeal, the Federal Circuit looked at the final result — the scope of Claim 6 has not changed. “Where the scope of claim 6 has not changed, there has not been improper claim broadening, and HP’s argument fails.”

The court went on to explain that the addition of those dependent claims would not have changed the scope of claim 6.

Thus, even were dependent claims 15 and 16 broader than unamended, independent claim 6, the remedy would not be to find claim 6 invalid as broadened, but to invalidate added claims 15 and 16 [for improper broadening].

Slip Op.  This decision sits well here, but will lead to further odd results when applied to ordinary prosecution because it allows for dependent claims that are broader than the independent version.

This post serves as a complement to Prof. Gugliuzza’s new remarks on a parallel case of In re Apple. – DC

by Dennis Crouch

In re Google (Fed. Cir. September 18, 2020) (Google III)

Google’s business pervades the lives of most Americans, including most citizens of the E.D. of Texas.  Google has millions of customers in the district; serves terabytes of data to, from, and within the district; and keeps detailed files on the activities of its citizens. Google also has lots of Texas lawyers.  Google is doing everything it can to move this case out of E.D.Texas.  The reality is though that Google doesn’t mind being in Texas, it just doesn’t want Texas style justice — where patent cases are on a direct path to a jury trial.

28 U.S.C. 1400(b) has a specific test for proper venue. An infringement lawsuit can only be filed in a district where either: (more…)

Guest Post by Prof. Paul R. Gugliuzza (Temple U.)

A rare thing happened at the Federal Circuit today. The court heard oral argument on a petition for a writ of mandamus. The petition was filed by the tech behemoth, Apple, in a patent infringement case filed against it in the Western District of Texas. In the petition, Apple seeks an order sending the case to the Northern District of California under 28 U.S.C. § 1404, which permits transfer “[f]or the convenience of parties and witnesses, in the interest of justice.”

Though transfer petitions are relatively common in patent cases, the Federal Circuit almost always decides them on the briefs alone. That the court scheduled oral argument—in a case arising out of the Western District of Texas, no less—has been interpreted as reflecting concern by the Federal Circuit about the judge shopping occurring in the Western District.

As Jonas Anderson and I showed in a recent Patently-O post and discuss in more detail in a draft article, the Western District’s case assignment rules permit plaintiffs to predict, with absolute certainty, which judge will hear their case. And plaintiffs are overwhelmingly choosing Judge Alan Albright, whose procedural rules and substantive decisions they find quite favorable.

That said, the Federal Circuit’s decision to hold oral argument on Apple’s petition could also reflect the fact that, in the midst of the COVID pandemic, it’s a pretty easy thing to do. For the past six months—and for the foreseeable future—the Federal Circuit has been conducting oral argument entirely by telephone. Indeed, that’s how I was able to listen to today’s arguments, live.

Before getting to a summary of that argument, some background about the case. The plaintiff is, like many plaintiffs in the Western District, a prolific non-practicing entity, Uniloc 2017 LLC. In September 2019, Uniloc sued Apple for infringing a patent on a system for controlling software updates.

Like more than 800 other patent cases over the past two years, Uniloc filed its case in the Waco Division of the Western District of Texas and—like 100% of cases filed in the Waco Division—it was assigned to Judge Albright. Apple sought transfer to the Northern District of California, noting that, out of 24 prior cases Uniloc had filed against it in the Eastern and Western Districts of Texas, 21 had been transferred.

But Judge Albright denied Apple’s motion in an order from the bench in May 2020. As covered here on PatentlyO, it took Judge Albright more than a month to issue an order explaining why he was doing so. When that order eventually issued, it noted, among other things, that Apple has stronger connections to the Western District of Texas than to the Eastern District and that the cases previously transferred out of the Western District (by Judge Lee Yeakel) were distinguishable because Apple’s activities in the Western District had grown significantly over the past couple years.

The Federal Circuit argument, it’s worth noting, wasn’t part of the court’s normal calendar of arguments, which typically take place during the first week of the month. Rather, it was the only case heard by a panel consisting of Chief Judge Prost, Judge Moore, and Judge Hughes.

Mel Bostwick, from Orrick, Herrington & Sutcliffe’s Washington, D.C., office, presented argument for Apple. In her view, the district court made two critical errors in denying transfer: First, it relied too heavily on the progress it had already made in the case as well as its already-scheduled trial date (which, under Judge Albright’s extremely speedy default schedule, is less than 18 months after the initial case management conference).

Second, according to Apple, the district court erred in applying the “cost to willing witnesses” factor in the transfer analysis. Though both Apple and Uniloc identified witnesses in California, Judge Albright, according to Apple, inappropriately discounted the relevance of those witnesses because they were willing to travel. But, Apple contended, the relevant question is the cost of their travel, not their willingness to do so.

Apple faced skeptical questioning from Judge Moore, who was, in fact, the only judge to ask a question of Apple until rebuttal. Judge Moore focused initially on the standard of review. To receive the extraordinary writ of mandamus, a party must show a “clear abuse of discretion” by the district judge. The fact that this case has some factual connection to the Western District—namely, Apple has a campus in Austin and a third party makes accused products in the district—seemed to raise doubts in Judge Moore’s mind about whether any error by the district court met that high bar.

Christian Hurt of the Davis Firm in Longview, Texas argued on behalf of Uniloc. He began by emphasizing the concerns about parties and witnesses located in the Western District that were initially raised by Judge Moore. Apple didn’t dispute, Uniloc noted, that it has an 8,000 employee campus in Austin, technical witnesses work there, and a third-party contractor makes accused products in the district.

Almost all the questions for Uniloc came from Chief Judge Prost. She asked about matters including: the exact location of the witnesses, whether it was clearly an abuse of discretion for the district court to rely on its progress and projected schedule in denying transfer, and whether Apple might have an alternative means of seeking relief, such as through a later mandamus petition or by seeking a stay pending related litigation elsewhere.

Toward the end of Uniloc’s argument, Judge Moore chimed in to ask whether, if the court found the district court had made errors in its transfer analysis, it would be appropriate for the Federal Circuit to vacate the decision and remand the case for further proceedings, rather than ordering transfer—a step the very same panel of Federal Circuit judges basically took in a  recent Western District case filed against the file storage company, Dropbox.

During Apple’s rebuttal argument, Judge Moore asked why transfer to California was warranted given the local interest in the case. Apple, Judge Moore observed, is one of the largest employers in the Western District—a far cry from the Eastern District, where Apple doesn’t even have stores anymore, for fear of aiding patent plaintiffs in establishing venue there. Judge Moore was unconvinced (to put it mildly) by Apple’s assertion that the local interest isn’t the interest of Western District of Texas and its residents, but the interest of “the people who created the accused technology,” in Cupertino.

*          *          *

So, what’s my take? The atmospherics are clearly troubling. There’s no doubt that Judge Albright is successfully courting patentees to file in his courtroom both by explicitly advertising to them and by adopting procedural rules and making substantive decisions that clearly favor them. But those larger dynamics, though they were discussed in Apple’s brief, weren’t even mentioned at oral argument. (Bostwick, Apple’s attorney, seemed to want to go there during rebuttal, but ran out of time.)

In this case, the Federal Circuit might struggle to find a legal justification for ordering transfer, particularly given high standard for mandamus. That said, the Federal Circuit rarely hesitated to transfer cases out of the Eastern District during its heyday as the nation’s patent litigation capital. In several cases, the Federal Circuit used the extraordinary writ of mandamus to engage in what seemed like pure error correction. It’s not out the question that the Federal Circuit would do something similar with the Western District, whether in this case or one of the other nearly 600 filed before Judge Albright this year alone.

Moreover, though the court competition and judge shopping that’s going on in the Western District is troubling, interlocutory appeals like the one Apple is pursuing can be costly and disruptive. That will be even more so if the Federal Circuit makes a habit of simply vacating orders denying transfer and remanding for further consideration, as Judge Moore suggested. The end result would be another round of briefing and argument—and possibly even discovery—on an issue entirely tangential to the merits of the case.

Whatever the outcome, this case between Apple and Uniloc shows how difficult it will be for the Federal Circuit, which can only hear the disputes that come before it, to change the systemic incentives that encourage judges to compete for patent cases and for plaintiffs to shop for those judges. As we suggest in our article, legislation or administrative rules mandating random case assignment and more particularly defining plaintiffs’ venue choices may be the only solution.

Paul Gugliuzza is Professor of Law at Temple University Beasley School of Law

USPTO Fees are changing at the end of the month. PCT Fees are changing on October 1, 2020; US National fees are changing on October 2, 2020. In general, the fees are going up, not down.  Beat the fees – file by September 30, 2020.

Bozeman Financial LLC v. Federal Reserve Bank of Atlanta, et al. (Supreme Court 2020)

The question in this case is whether the Federal Reserve Banks are people.  The Patent Act allows any “person” to file a petition for covered-business-method review (or IPR/PGR). Return Mail, Inc. v. United States Postal Serv., 139 S. Ct. 1853 (2019) held (1) the U.S. Gov’t is not a person under the statute and (2) consequently the USPS (a branch of the US gov’t) is not permitted to petition the USPTO for review of a patent.  The question before the Supreme Court is whether these banks are part of the government.

Whether the regional Federal Reserve Banks—the “operating arms” of the Federal Reserve System, which is the central bank of the United States—are “distinct” from the Federal Government, and qualify as “persons” permitted to seek post-issuance patent review under the America Invents Act, when the Federal Government may not under the Court’s holding in Return Mail, Inc. v. United States Postal Serv., 139 S. Ct. 1853 (2019).

Question presented.

Alexander Hamilton was instrumental in the creation of the First Bank of the United States.  That Bank’s charter ended in 1811, but the foundation served for future national banks and eventually for creation of the Federal Reserve in 1913. The Federal Reserve system includes twelve regional Federal Reserve Banks that are largely controlled by private banking interests. The twelve are self-described “instrumentalities of the United States that, collectively, make up the operating arm of the Federal Reserve System, the central bank of the United States.” (Bank Complaint).  The system as a whole is controlled by the Board of Governors of the Federal Reserve System. The Board members are presidential appointees.

Bill Bozeman’s patents cover what he calls “Universal Positive Pay” for fraud detection and check clearing. Back in 2017, the 12 Federal Reserve Banks (but not the Board) sued Bozeman seeking a declaratory judgment of non-infringement.  The banks then also filed for Covered Business Method (CBM) review of the patents at the USPTO.  The PTO instituted review and concluded that the claims were ineligible under Alice Corp. Pty. Ltd. v. CLS Bank Int’l, 573 U.S. 208 (2014).  In its decision, the Federal Circuit affirmed and also held that the banks are “persons” under the statute because they “are distinct from the government for purposes of the AIA.”

In other contexts, courts have found a very close link between the Banks and the Gov’t.

• Fed. Reserve Bank of Bos. v. Comm’r of Corps. & Taxation of Com. of Mass., 499 F.2d 60, 62 (1st Cir. 1974) (Bank is a “public governmental body” whose “interests seem indistinguishable from those of the sovereign”)
• Jet Courier Servs., Inc. v. Fed. Reserve Bank of Atlanta, 713 F.2d 1221 (6th Cir. 1983) (Banks are not “persons” under the Sherman Act because they are part of the Federal Reserve System, “an agency of the federal government.”)
• Schroder v. Volcker, 864 F.2d 97, 99 (10th Cir. 1988) (no antitrust action against defendants “affiliated with the Federal Reserve System”, including individual banks)
• United States v. Hollingshead, 672 F.2d 751 (9th Cir. 1982) (Fed Reserve Bank employees are public officials for purposes of anti-bribery statute).
• Berini v. Fed. Reserve Bank of St. Louis, 420 F. Supp. 2d 1021, 1028 (E.D. Mo. 2005) (“[C]ontrol and supervision of the federal reserve banks is vested in a Board of Governors appointed by the President with the advice and consent of the Senate”)

This case is not huge for the patent system — although there are hundreds of federally-created entities that might be “people.”  In addition, the CBM program has sunset and is unlikely to be revived.

The case is still a big deal as our country discusses the role of socialist governmental policies providing a safety net for Americans. The US system is already ripe with “private” entities designed to serve a public good: Federal Reserve banking system; Fannie Mae; Freddie Mac; Highly regulated utilities (that are given the power of eminent domain); etc.  For over 100 years, this approach has been a form of back-door socialism that becomes palatable because of paperwork showing a separation from government. This case would shine some interesting light on the field with the simple question — Are the Federal Reserve Banks part of the U.S. Government?

by Dennis Crouch

Once a patent issues, it is presumed valid. “A patent shall be presumed valid.” 35 U.S.C. 282(a). In patent litigation, this has traditionally meant that a complaint for infringement need not re-establish the patent’s validity. Rather, validity challenges arise as affirmative defenses as part of the answer.

That traditional approach is no longer followed by the courts in the Post-Alice patent eligibility era.  Courts now regularly dismiss patent cases upon finding that the patentee failed to state a claim because the patent is invalid under Section 101.  In response to that potential, patentees are have begun preemptively bulking-up their complaints with factual allegations to support the patent’s validity.

A new petition for writ of certiorari in Whitserve LLC v. Donuts Inc. (2020) highlights this issue.  Back in 2018, Whitserve sued Donuts for infringing the claims of its two patents covering a method for managing due-date reminders for clients of professional-services.  U.S. Patent Nos. 5,895,468 and 6,182,078.  On a R. 12(b)(6) motion, the district court dismissed the complaint – finding that the patent was invalid as a matter of law and that – therefore – the complaint failed to state a legally cognizable claim for relief. On appeal, the Federal Circuit affirmed – explaining again that “patent eligibility can be determined at the Rule 12(b)(6) stage if there are no plausible factual allegations to impede such a resolution.” (Quoting Aatrix).  The suggestion here is that plaintiffs really do need to be making their validity case within the pleadings.

Now, the case is up before the Supreme Court on Whitserve’s recently filed petition. Question presented:

If a patentee makes factual assertions that its claimed invention is directed to patent eligible subject matter under 35 U.S.C. § 101, including assertions that the claimed invention does not consist of well understood, routine, or  conventional activity and that the claimed invention is supported by evidence of commercial success, is a district court permitted to overlook the patentee’s assertions, find that the claimed invention is directed to patent ineligible subject matter, and dismiss the patentee’s complaint under Rule 12(b)(6) given the requirements of Rule 12(b)(6) analyses and the statutory presumption of § 282(a)?

[Petition].

When I read the question presented, I instantly wanted to make some amendments — in particular, I wanted to focus not on bar “factual assertions that” but rather on “non-conclusory factual assertions showing that patent is plausibly eligible.”  My transformation focuses on the plausibility standard from Iqbal and Twombly.  In those cases, the Supreme Court raised the standard for “showing that the pleader is entitled to relief” under R.8(a).  On my second time through, however, I began to really question this approach.

We are talking here about a pleading that attempts to preempt a potential affirmative defense — normally the plaintiff does not even need to plead a response to an affirmative defense.

[E]ven after the defendant has pleaded an affirmative defense, the federal rules impose on the plaintiff no obligation to file a responsive pleading.

Fernandez v. Clean House LLC, 883 F.3d 1296 (10th Cir. 2018).  And, when a Reply to an affirmative defense is ordered, it is sufficient to simply deny the allegations of the defense rather than explain or offer competing factual allegations.  In that situation, the non-conclusory / plausibility standard of Iqbal does not apply.

All this leads me to say that – for 12(b)(6) purposes, even conclusory factual allegations regarding eligibility may be sufficient to overcome a motion to dismiss. Of course, at that point, the court can jump quickly to a R.56 Summary Judgment question — allowing special early discovery on the eligibility issue to see whether there is any evidence to support the allegations.

* Note, the image above comes from a design patent owned by Krispy Kreme parent company HDN Development.

Is anyone working hard on a Covered-Business-Method (CBM) Review petition?  The program is ending soon. I believe the last day to file your petition is September 15, 2020.

by Dennis Crouch

I previously wrote about the case of Bank v. Al Johnson’s Swedish Rest., Docket No. 19-01880 (Fed. Cir. 2019).  The dispute is over whether the USPTO should cancel Al Johnson’s registered trademark for goats on a green roof. To be clear – the mark is not the image of goats on a roof, but instead is an actual building with live goats walking around on the roof. [Goat Cam]

Bank challenged the registration on several grounds, including improper functionality and disparaging (toward the goats and their human friends).  The problem in the case for Bank is that he is not a competitor or customer. Banks is not injured by the mark in any concrete way other than being offended by its existence.

The TTAB dismissed the opposition for lack of standing. That decision was then affirmed on appeal since Bank provided neither a real interest nor a reasonable basis for his belief of damage.  The court noted that the “offense” injury was substantially undermined by Tam.

• Discussion of the Federal Circuit Decision;
• Discussion of the Supreme Court Petition, subsequently denied.

In its original decision, the Federal Circuit also awarded attorney fees to Swedish Restaurant under Fed. R. App. P. 38:

If a court of appeals determines that an appeal is frivolous, it may, after a separately filed motion or notice from the court and reasonable opportunity to respond, award just damages and single or double costs to the appellee.

Id. The court found that the Banks appeal was frivolous. Banks is an attorney and represented himself in the appeal. Usually pro se parties are given more leeway because of their lack of training and experience in the system. However, an attorney representing himself is not given such leeway:

Even though Mr. Bank appears pro se before us, he is an attorney and bears the commensurate obligations. Accordingly, we grant Swedish Restaurant’s motion for costs and attorney fees, including the costs and fees incurred in relation to the parties’ sanctions motions, and
deny Mr. Bank’s motion for sanctions.

Costs and attorney fees to Swedish Restaurant.

Federal Circuit Original Opinion.

Following the court’s decision, there was some debate on attorney fees. In particular, Swedish Restaurant requested that the court clerk enter the attorney fee award. Banks protested — arguing that attorney fee awards must be calculated and awarded by the court, not the clerk.  The Federal Circuit agreed on that point and today awarded all of Swedish Restaurant’s requested fees of $28,523.00.  (The Clerk separately taxed the costs at $241.54.)

by Dennis Crouch

The decision in this case is short and non-precedential, but raises an interesting Arthrex issue on Unconstitutional Appointments. 

Snyders Heart Valve LLC v. St. Jude Medical, LLC (Fed. Cir. 2020) (SnydersStJude)

Snyders’ US Patent No. 6,821,297 covers a collapsible artificial heart valve that can be attached through a blood vessel rather than open-heart surgery.

After being sued for infringement, St. Jude filed an IPR petition and won a determination that the claims would-have-been-obvious at the time of the invention. Snyders appealed.

Rather than addressing the merits of the obviousness case, the Federal Circuit has followed its precedent set by Arthrex and has has vacated the PTAB’s judgment because the judges were appointed in an unconstitutional manner. On remand, a reconstituted–and now magically constitutional–PTAB panel will re-do the trial (unless the Supreme Court intervenes before then).

In its argument, Snyders suggested that it should receive some particular treatment from the court because of a potential conflict of interest with Dir. Iancu.  Prior to joining the USPTO, Iancu was in private practice and represented St. Jude in a parallel proceedings.  Although Dir. Iancu has recused himself from the case, Snyders argues that the Director’s conflicts are not so easily erased.  Rather, an attorneys conflicts regularly extend to subordinate employees as well.  Here, the Arthrex remedy comes into play because the court in that case gave more direct authority supervisory to the PTO Director. The following argument comes from Snyders’ brief:

The concept that disqualification of an attorney may extend to that attorney’s subordinate employees is well established. For example, the American Bar Association’s Model Rules of Professional Conduct recognize a conflict where representation of a client is materially limited by an attorney’s personal interest. See Model Rules of Prof’l Conduct R. 1.7(a)(2) (2016). Those rules also recognize that disqualification of an attorney due to a personal conflict may be imputed to fellow employees where the employees would be materially limited due to their loyalty to the attorney.

Snyders Brief.

The Federal Circuit found the argument here “without merit . . . the Deputy Director’s role sufficiently removes any potential taint of the Director’s conflict.”  The Court did not address the particular issue here regarding the heightened supervisory authority of PTAB judges coming-out-of Arthrex.

by Dennis Crouch

Thousands of patents claims have been cancelled by the PTAB in inter partes review proceedings.  These are cases where a third party was willing to pay hundreds-of-thousands-of-dollars to cancel a set of claims.  A good number of those patents would have been enforced by a district court. In fact, a good number of those patents have been enforced in Federal Court.

Personal Audio, LLC v. CBS Corporation (Supreme Court 2020) again builds a montage of competing patent tribunals — a district court that enforces and an administrative court that undermines.  This case adds an important third tribunal to the story — the jury.  The disappointing climax is usually the same — spent on the Federal Circuit who sides with the administration.

In this case, the jury sided with Personal Audio — finding the asserted claims of its US8112504 both valid and infringed. (Verdict excerpt below from September 2014).  The jury awarded $1.3 million.

By the time of the jury verdict, the IPR petition filed by the non-profit EFF had already been instituted by the PTAB. Tthe district court allowed post-judgment motions to flounder for months and by April, the PTAB had released its final judgment finding the same claims invalid.  (Note, the PTO no longer institutes IPRs on patents that are close to trial).

The IPR was affirmed on appeal, and the district court relied upon that determination to cancel the jury verdict and enter a final judgment in favor of the Defendant.  The Federal Circuit then affirmed based upon its prior precedent of XY, LLC v. Trans Ova Genetics, 890 F.3d 1282, 1294 (Fed. Cir. 2018); Dow Chemical Co. v. Nova Chemicals Corp. (Canada), 803 F.3d 620, 628 (Fed. Cir. 2015); ePlus, Inc. v. Lawson Software, Inc., 789 F.3d 1349, 1358 (Fed. Cir. 2015); Fresenius USA, Inc. v. Baxter Int’l, Inc., 721 F.3d 1330 (Fed. Cir. 2013).

Now, Personal Audio is up before the U.S. Supreme Court with a few interesting questions. The basic argument is that collateral estoppel (issue preclusion) is not an open-and-shut business. Rather, any time a prior judgment of invalidity is invoked, the “patentee-plaintiff must be permitted to demonstrate, if he can, that he did not have a fair opportunity procedurally, substantively and evidentially to pursue his claim the first time.” Blonder-Tongue Laboratories, Inc. v. U. of Illinois Found., 402 U.S. 313, 332–33 (1971).  The Restatement of Judgments § 29 further walks through a series of reasons and factors for issue preclusion might not apply to a later case.

Blonder-Tongue was a very big case for defendants supporting non-mutual issue preclusion. The point here, though is that it set limits and provided a contextual analysis while the Federal Circuit has moved to an if-then rule-based approach post-IPR.

Here, the patentee argues against estoppel because (1) the PTAB judges were unconstitionally appointed (via Arthrex); and (2) the PTAB determination overturns a prior jury’s finding of facts in a case protected by the 7th Amendment Right to Trial.

The trouble – as usual – is that the claims are pretty bad: Personal Audio’s patent claims a system for distributing “a series of episodes represented by media files via the Internet.”  The new portion appears to be updating a compilation file of currently available episodes.  The list of new episodes (and their links) are downloaded by a client computer that can then use the links to request the actual media files.

The other big difficulty here is a question of when does the jury verdict stick.  Is it worthless until the judge enters final judgment?

The Supreme Court will rule on the petition later this Fall.

by Dennis Crouch

Baxalta Inc. v. Genentech, Inc., 2020 WL 5048435 (Fed. Cir. Aug. 27, 2020)

Although not en banc, this case was picked-up by five different Federal Circuit judges with two separate opinions and one recusal.

Sitting by designation as a D.Del. district court, Judge Dyk sided with the accused infringer Genentech by narrowly construing Baxalta’s US7033590.  Although perhaps most jury members would need some help with definitions, the disputed terms are incredibly basic and fundamental to anyone involved with biotech. Disputed terms:

• Antibody; and
• Antibody fragment.

Following claim construction, the parties stipulated to non-infringement.

On appeal, the Judges Moore, Wallach, and Stoll took up the claim construction question.  However, after oral arguments Judge Stoll recused herself and Judge Plager stepped-in.  The new panel rejected Judge Dyk’s claim construction and consequently vacated the non-infringement judgment.

The patent is directed to a treatment for hemophilia and claims “an isolated antibody or antibody fragment” that binds to FactorIXa and increased its procoagulant activity.

Judge Dyk explained that the term antibody might have various potential meanings.  But here, the patentee expressed a narrower definition within its specification with the following statement:

Antibodies are immunoglobulin molecules having a specific amino acid sequence which only bind to antigens that induce their synthesis. . . . Each immunoglobulin molecule consists of two types of polypeptide chains. Each molecule consists of large, identical heavy chains (H chains) and two light, also identical chains (L chains).

‘590 at Col. 5.  Applying this portion of the patent, the district court limited the scope of antibody to molecules having one set of identical H-chains and a set of identical L-chains.

Genentech’s product uses an antibody organized with a variety of H- and L- chains (not all the H-chains are identical to one another). Thus, Judge Dyk’s requirement of identical chains led to the admission of non-infringement.

On appeal, the Federal Circuit took Judge Dyk to task — finding that the patent did not suggest such a narrow definition of the term. Dependent claim 4, for instance, identifies the antibody as being selected from “the group consisting of … a chimeric antibody, a humanized antibody, … [and] a bispecific antibody.”  The court explains that none of these antibody types in the dependent claim conform with the narrow definition of the term as used in claim 1.

The district court’s construction which excludes these explicitly claimed embodiments is inconsistent with the plain language of the claims. . . . The plain language of these dependent claims weighs heavily in favor of adopting Baxalta’s broader claim construction.

Slip Op.  The court then looked at the definition discussed by Judge Dyk and concluded that it was a “generalized introduction” rather than a “definitional statement.” In this process, the court noted the lack of patent-profanity such as “the present invention is … ”  The court’s conclusion is supported by other aspects of the specification that refers to the bispecific and other forms of antibodies that do not conform to Judge Dyk’s definition. The result: “The claim construction excluding these disclosed and claimed embodiments is therefore incorrect.”

The case also included some prosecution history. In particular, the patentee amended its claims from “antibody derivative” to “antibody fragment.” On appeal, the court found no clear statement from the prosecution history regarding how that amendment modified the scope. Without a ‘clear statement’ there is no prosecution disclaimer leading to disavowal of scope.

by Dennis Crouch

Facebook v. Windy City (Fed. Cir. 2020) [WindyCity Original OP] [WindyCity Revised OP][WindyCity Mark-up Showing Changes]

This was an important PTAB JOINDER case when it was decided by the Federal Circuit back in March 2020.  (Panel of Chief Judge Prost with Judges Plager and O’Malley).  The en banc court has now denied rehearing, but the original panel has also revised its opinion to take into account the the Supreme Court’s intervening decision in Thryv.

In Civil Procedure, joinder is typically divided into distinct area: joinder of claims and joinder of parties.  The AIA statute provides only for joinder of parties, but the PTAB allowed Facebook to join itself to a pending IPR in order to add additional claims. Facebook had wanted to do this type of joinder, because its later-filed IPR petition would have otherwise been time-barred under § 315(b) — and the patent act particularly notes that the 1-year 315(b) time-bar “not apply to a request for joinder”.  In its first decision, the appellate panel held that the self-joinder approach was not permitted by the statute.

Since March, the Supreme Court decided Thryv — and overturned a set of Federal Circuit cases reviewing various petition-stage decisions by the PTAB.  In Thryv, the court held that the USPTO’s interpretation of the § 315(b) one-year time-bar was not reviewable on appeal based upon the “no appeal” provision of § 314(d).

Now the original panel has revised its opinion in Windy City to take into account Thryv.  In its decision, the court explained that 315(c) joinder requires two decisions: (1) determination of whether the joinder-applicant’s petition warrants institution (including whether it is time-barred); and (2) determination of whether joinder should be allowed.  Under Thryv, the first portion of the test is not reviewable on appeal, but the second portion is appealable.

The statute requires the Director (or the Board on behalf of the Director) to make a “joinder decision.”  The statute makes clear that the joinder decision is made after a determination that a petition warrants institution, thereby affecting the manner in which an IPR will proceed.

Thus, the joinder decision is a separate and subsequent decision to the intuition decision.  Nothing in § 314(d), nor any other statute, overcomes the strong presumption that we have jurisdiction to review that joinder decision. . . .

Accordingly, we have jurisdiction to review the Board’s joinder decisions in this case to determine whether the Board (on behalf of the Director) acted outside any statutory limits under § 315(c).

Revised Opinion. Thus, the holding:

Because joinder of the new claims was improper, we vacate the Board’s final written decisions as to those claims, but because we lack authority to review the Board’s institution of the two late-filed petitions, we remand to the Board to consider whether the termination of those proceedings finally resolves them.

Id. I expect that on remand, the PTO will dismiss the late-filed petitions.

In its revised opinion, the court maintained its unanimous secondary opinion offering “additional views” on deference to Precedential Opinion Panels at the PTO.

Guest post by Saurabh Vishnubhakat, Professor at the Texas A&M University School of Law and College of Engineering.  Professor Vishnubhakat was formerly an advisor at the USPTO, but his arguments here should not be imputed to the USPTO or to any other organization.

This week, four iconic Silicon Valley technology companies—Apple, Cisco, Google, and Intel—sued the USPTO under the Administrative Procedure Act.  The lawsuit challenges the USPTO’s so-called NHK-Fintiv rule, named after a pair of inter partes review decisions in the PTAB that the agency previously designated as precedential.

The 23-page complaint, docketed as Case No. 5:20-cv-06128 in the Northern District of California, is worth reading in full.  Yet what is especially striking about the lawsuit, and worth considering more deeply, is a particular pair of arguments at the heart of the challenge.  One is that the NHK-Fintiv rule is contrary to the policy and text of the AIA and therefore exceeds the Director’s authority.  The other is that the NHK-Fintiv rule is procedurally infirm because it was not promulgated through APA notice-and-comment rulemaking.

The NHK-Fintiv Rule

The disputed USPTO policy allows the PTAB to deny institution of an inter partes review petition based on how far a parallel U.S. district court proceeding on the same patent has already gone.

NHK Spring v. Intri-Plex

The policy was first articulated in NHK Spring Co. v. Intri-Plex Techs., No. IPR2018-00752, Paper 8 (Sept. 12, 2018).  There, a panel of the PTAB declined to institute NHK Spring’s petition against an Intri-Plex patent where a parallel infringement suit was already pending between the same parties in the Northern District of California.

In denying institution, the panel cited its discretion under 35 U.S.C. § 325(d) as well as under § 314(a).  First came § 325(d), which empowers the Director to “determine the manner in which the post-grant review or other proceeding or matter may proceed, including providing for the stay, transfer, consolidation, or termination of any such matter or proceeding.”  Here, the panel applied the nonexclusive factors of the PTAB’s prior informative opinion in Becton, Dickinson and concluded that the art and arguments now asserted in the PTAB were already considered (and overcome) during examination.

Though it found this analysis sufficient on its own, the panel then also went on to exercise its discretion under § 314(a), which makes a “reasonable likelihood” of invalidating at least 1 of the challenged claims a necessary—but not sufficient—condition for instituting review.  Where review is permissible, the Director may still decide in his discretion to deny review, and the NHK panel found it compelling that the parallel proceeding in U.S. district court was “nearing its final stages”—with a five-day jury trial already set for six months before the PTAB’s own proceeding would conclude.

The principle of NHK—that the “the advanced state of the district court proceeding is an additional factor that weighs in favor of denying the Petition under § 314(a)”—forms the first part of the policy now being challenged.

Apple v. Fintiv

That policy was further elaborated in Apple Inc. v. Fintiv, Inc., No. IPR2020-00019, Paper 11 (Mar. 20, 2020).  There, a panel of the PTAB ordered supplemental briefing at the institution stage of Apple’s petition against a Fintiv patent where a parallel infringement suit was pending between the same parties in the Western District of Texas.  Fintiv had already argued in its preliminary response that the “advanced state” of the parallel proceeding warranted discretionary denial under NHK, as the same issues were before the district court and trial there had already been set.  (The setting of a trial date had come after Apple’s petition but before Fintiv’s response, making additional briefing appropriate.)

The panel then set out a number of factors to consider when evaluating whether the state of a parallel proceeding warrants discretionary denial under NHK:

• whether the court granted a stay or evidence exists that one may be granted if a proceeding is instituted;
• proximity of the court’s trial date to the Board’s projected statutory deadline for a final written decision;
• investment in the parallel proceeding by the court and the parties;
• overlap between issues raised in the petition and in the parallel proceeding;
• whether the petitioner and the defendant in the parallel proceeding are the same party; and
• other circumstances that impact the Board’s exercise of discretion, including the merits.

NHK was designated as precedential in May 2019 and Fintiv in May 2020.  Taken together, the NHK-Fintiv rule represents a policy of denying institution where a parallel district court proceeding is so far along and so substantially similar in art and argumentation that it would be best to conserve USPTO resources rather than undertake a largely or entirely duplicative review.

The Policy and Text of the AIA

That policy choice is firmly rejected in the opening argument of the APA challenge.  The plaintiffs identify the inter partes review system as a “centerpiece of Congress’s efforts to strengthen the U.S. patent system” through post-grant error correction.  By their account, the system of PTAB adjudcation responded to an environment where “questionable patents were too easily obtained and too difficult to challenge through existing procedures”—and, indeed, this language aptly cites the AIA House Judiciary Committee Report.  Thus, to deny institution under the NHK-Fintiv framework weakens the very purpose of PTAB review through artificial limits that are “found nowhere in the AIA.”  However, though there is much to agree with in the line of argument that follows, it suffers from at least two important weaknesses.

PTAB Review as an Alternative to the Courts

One weakness is that while the PTAB is desirable over the baseline of Article III courts, there are important and under-appreciated limits to this desirability.  It is certainly true that the PTAB was intended as “an improved alternative to litigation” on questions of patent validity.  Indeed, my coauthors and I have similarly argued that the PTAB offers a number of important advantages over the Article III courts, including lower barriers to standing, lower cost, lower delay, and lower rates of error.

However, it does not follow that PTAB review remains preferable regardless of what happens in the Article III courts.  By the time court litigation has reached a stage advanced enough that the NHK-Fintiv doctrine would apply, much of the cost of litigation has already been sunk, especially by the close of discovery and the scheduling of trial, as in NHK itself.  Meanwhile, the problem of delay is turned on its head, as it is the court that will now likely finish before the PTAB would.  The problem of Article III standing is largely irrelevant, as the defendant can point not merely to the threat of suit but the actual suit itself.  Much, though not all, of the marginal benefit from PTAB review relative to the federal courts is already dissipated.

Moreover, while it is true that decision making in the PTAB is done by administrative judges who have relevant technical as well as legal expertise, this benefit is also dissipated to some degree by a late-stage federal court proceeding.  By that time, considerable effort and investment has already been sunk into educating the judge or jury.  This, after all, is where much of the cost of litigation goes, and what makes expert administrative judges an attractive value proposition is that they do not require nearly so much education in each case.  The more that such investments have been made anyway, the less that PTAB review is a clear cost-saving.

Finally, there is the problem inherent to error correction, a problem starkly highlighted by the legislative design of PTAB review.  It is true, as the plaintiffs point out, that “while bad patents can be held unpatentable in IPR by a preponderance of the evidence . . . those same patents will survive litigation unless the challenger proves invalidity by clear and convincing evidence.”  But as I have pointed out in testimony before the FTC, the same is also true of good patents—and there is no way to distinguish the good from the bad up front.  If there were, error correction itself would be unnecessary.

Agency Discretion to Deny Review

The second weakness is that agency discretion carries not only significant structural benefits when protecting agreeable outcomes but also substantial obstacles when the outcomes go the other way.  The crux of the case against discretionary denials under the NHK-Fintiv rule is that “no provision in the AIA expressly requires or even permits the Director (or the Board as his delegee) to deny IPR petitions based on pending litigation involving the same patent claims.”

It is repeated throughout the argument, too, that the principle of NHK and the additional factors enumerated in Fintiv are to be found “nowhere in the AIA” and are, for that reason, outside the Director’s authority.  In this telling, what discretion the Director does have is limited to § 325(d), which is concerned more with managing multiple proceedings inside the agency itself than with doing so across an interbranch court-agency divide.  This matters because the NHK-Fintiv framework is an elaboration of institution authority specifically under § 314(a), not of case management authority under § 325(d).

However, the principle that § 314(a) gives the Director discretion—broad discretion—to deny otherwise meritorious petitions is, by now, fairly well established in Federal Circuit and Supreme Court case law.  For example, the two most significant cases involving the judicial unreviewability of the Director’s institution power—Cuozzo v. Lee in 2016 and Thryv v. Click-to-Call earlier this year—take just this view.  The Court in Cuozzo held explicitly that “the agency’s decision to deny a petition is a matter committed to the Patent Office’s discretion” and cited § 314(a) with an explanatory parenthetical that there is “no mandate to institute review.”  Likewise, the Court in Thryv expanded the scope of that unreviewable discretion to include conditions on institution as well—there, the condition in dispute was the one-year time bar of 35 U.S.C. § 315(b).

Indeed, some of the plaintiffs who now seek to cabin the Director’s institution-related discretion previously endorsed those very same positions before the Court.  In Cuozzo, Apple submitted a brief as amicus curiae supporting the USPTO Director’s assertion of unreviewable discretion in matters of institution.  Intel did the same in Thryv, arguing that a “decision not to institute review is committed to agency discretion.”

These complications in the case against the NHK-Fintiv rule cast serious doubt on the view that the policy choices embodied in that rule contradict the AIA.

The Choice of Rulemaking vs. Adjudication

Beyond the substance of the USPTO’s policy of sometimes denying inter partes review based on the status of parallel court litigation, there also lies an alternative argument that the policy is procedurally defective.  Here, the challenge springs from the familiar APA values of public input and transparency, which are traditionally accomplished by notice-and-comment rulemaking.  By contrast, the disputed USPTO policy was adopted “by designating the NHK and Fintiv decisions as precedential through a unilateral, internal process that involved no opportunity for public comment and no consideration by the Director of any public input.”

In this regard, the challenge to NHK-Fintiv certainly has merit as a matter of desirable administrative practice, but it is not at all clear that this makes the USPTO’s approach legally deficient.  For over 70 years, the Supreme Court has left the form of policymaking up to agencies themselves.  The Court’s 1947 opinion in SEC v. Chenery Corp. (Chenery II) explained that “the choice made between proceeding by general rule or by individual, ad hoc litigation is one that lies primarily in the informed discretion of the administrative agency.”

In Chenery II, the Court recognized that “problems may arise in a case which the administrative agency could not reasonably foresee, problems which must be solved despite the absence of a relevant general rule.”  The Court also pointed to a touchstone of agency practice that is especially important to the PTAB—expertise—and noted that “the agency may not have had sufficient experience with a particular problem to warrant rigidifying its tentative judgment into a hard and fast rule.”  Both of these considerations point to adjudication as an acceptable mechanism for making policy.

To be sure, rulemaking offers significant benefits, and not only the public participation and transparency that the present APA challenge cites.  Rulemaking also fosters greater predictability, both by specifying rules more fully in advance and by raising the agency’s own political costs from visibly changing course.  For these reasons, I have argued in my own recent work about Patent Office policymaking that setting PTAB policy through rulemaking, such as the USPTO’s 2018 change to the PTAB’s claim construction standard, will often be preferable to shifting and incrementalist adjudications.

Still, a well advised preference is not the same thing as a legal requirement.  Moreover, though notice-and-comment rulemaking is a direct and well established way of securing public input, it’s not as if PTAB adjudication does not allow for meaningful public input.  The plaintiffs themselves note that members of the public are entitled to nominate PTAB opinions for designation as precedential.

Beyond this, the PTAB has accepted and, at times, even invited amicus curiae briefs in cases of public importance, such as whether tribal sovereign immunity defeats inter partes review.  Similarly, under the current USPTO Standard Operating Procedure No. 2 (Rev. 10), cases before the agency’s Precedential Opinion Panel may also be opened for amicus curiae briefing.  In short, where the USPTO exercises its prerogative to make policy through adjudication, it need not ignore public input to do so.

Conclusion

The APA challenge to the NHK-Fintiv rule, like much of the USPTO’s own recent policymaking, balances a range of important considerations and reaches a position that is coherent and reasonable.  The weakness—if it can be called that—of the challenge is that it represents merely one reasonable position among several, especially given the Supreme Court’s views on agency discretion in general and USPTO discretion in particular.  If the challenge eventually fails to dislodge the disputed policy, then the reason will likely be that, like most agencies, the USPTO enjoys wide latitude that is difficult to paint as unreasonable.