First Circuit

Antitrust treatment of FDA ‘Orange Book’ listings

In In re Lantus Direct Purchaser Antitrust Litigation,^[1] the First Circuit held that a drug manufacturer could be held liable under the antitrust laws for allegedly misusing the regulatory process to delay competition when it improperly submitted a patent to the US Food and Drug Administration (FDA) in connection with its drug product, which had the effect of triggering an automatic 30-month stay on the final approval of applications to market competing drugs. But significantly, the court also recognized that a manufacturer in this context may assert a defense to liability by showing that it had a reasonable, good-faith belief that it was required to submit its patent to the FDA under the applicable statutory and regulatory provisions.

Understanding the issues presented in the case requires a brief overview of the framework for drug approvals, and the manner in which patent protections intersect with the regulatory approval process. When a drug manufacturer files an application with the FDA seeking approval of a new drug, it must include a list of patents that ‘claim[] the drug for which the applicant submitted the application or which claims a method of using such drug with respect to which a claim of patent infringement could reasonably be asserted.’^[2] The FDA maintains the list of patents in a publication called Approved Drug Products with Therapeutic Equivalence Evaluations, commonly known as the Orange Book. But the FDA does not ‘review the patent information for its accuracy and relevance,’ leaving that responsibility with the manufacturer itself.^[3] The listing of a patent can have significant consequences for competition (especially from manufacturers of generic drugs), because companies seeking to file follow-on abbreviated applications that reference the studies used for the original, already-approved drug must file a certification to the listed patents.^[4] Specifically, as to listed patents that have not yet expired, the second applicant generally must either agree to delay its own approval until the relevant patents expire, or it can file a Paragraph IV certification asserting that the patents are invalid or would not be infringed by its product.^[5] If a Paragraph IV certification is filed, then the manufacturer of the first-approved brand product may file an immediate suit for patent infringement, which generally results in an automatic 30-month stay, during which the FDA cannot approve the new application.^[6]

This case concerns an antitrust challenge to a patent listing decision made by the defendant, Sanofi GmbH. Sanofi is the manufacturer of insulin glargine (brand name, Lantus®), which is a long-lasting form of insulin used to manage diabetes. Sanofi also received FDA approval to market insulin glargine in a disposable injector pen device called the Lantus SoloSTAR. In 2013, Sanofi submitted to the FDA a patent for listing in the Orange Book in connection with Lantus SoloSTAR. As a result, when competitors later submitted applications to the FDA to market insulin glargine in injector pens, they were required to file patent certifications, leading to automatic 30-month stays of approval. The plaintiffs in this case, a putative class of direct purchasers of insulin glargine products, allege (among other things) that Sanofi’s patent listing was improper, because the relevant patent (which related only to a component of the SoloSTAR device) did not qualify as a patent that claims either the relevant drug or drug product (i.e., insulin glargine or Lantus SoloSTAR). The plaintiffs further allege that Sanofi possessed monopoly power in the relevant market, and that the improper listing by Sanofi of its patent in the Orange Book constituted unlawful monopoly conduct that suppressed competition.

For the purposes of the motion to dismiss, Sanofi did not dispute the issue of monopoly power, focusing instead only on whether submission of its patent could qualify as monopoly conduct. The district court dismissed the plaintiffs’ claims, reasoning that Sanofi’s decision to list the patent was reasonable and could not be considered ‘objectively baseless’, given ambiguities in the FDA’s listing requirements.^[7] However, the First Circuit reversed and remanded for further proceedings.^[8]

To start, the First Circuit concluded that although the relevant FDA listing requirements are complex, they unambiguously do not support Sanofi’s decision to list a patent associated only with a component of SoloSTAR, where the patent does not claim either the relevant drug (insulin glargine), SoloSTAR itself, or a method of using either. The First Circuit also concluded, however, that even though a section 2 case does not typically examine the intent of monopolist, a different rule was required where the alleged anticompetitive conduct stemmed from the misapplication of FDA regulatory requirements, since good-faith errors should not result in automatic antitrust liability. On that point, the First Circuit noted that an improper failure to list a patent in the Orange Book can also have competitive consequences by depriving potential competitors of notice and certain procedural benefits; indeed, the plaintiffs have filed antitrust claims premised on a failure to list. The First Circuit thus took the middle ground. Borrowing from case law involving antitrust law and submissions to the Federal Communications Commission, the First Circuit held that Sanofi could raise a defense to antitrust liability by showing that its actions were the result of a reasonable, good-faith attempt to comply with its regulatory obligations to the FDA. The First Circuit remanded to allow further factual development of this defense.

Finally, the First Circuit rejected Sanofi’s separate argument that the plaintiffs had not adequately alleged antitrust causation. Sanofi contended that the automatic 30-month stay resulting from the Orange Book listing did not extend its monopoly because it ultimately entered settlements with its potential competitors that resulted in licensed entry dates well after the 30-month stays expired. However, the First Circuit concluded that, at least at the motion to dismiss stage, it was plausible to infer that the automatic 30-month freeze on FDA approval of other companies’ products may have affected how litigation ensued and thus could have affected competition.

This case has continued to proceed in district court on remand from the First Circuit. In one notable decision, the district court denied a motion to dismiss certain claims brought by a direct purchaser, FWK Holdings, LLC (FWK).^[9] FWK brought its claims based on an assignment purchased in bankruptcy from Frank W Kerr Company. Sanofi argued that because Kerr made its last purchase in May 2016, FWK lacked Article III standing to pursue claims on behalf of absent class members based on alleged anticompetitive conduct that occurred after Kerr’s purchases had ceased. The district court determined that FWK could pursue these claims because both Kerr and the absent class members are alleged to have suffered the same type of injury based on Sanofi’s alleged continuing scheme to block competition in the market for insulin glargine and thus shared an interest in litigating the claims arising from Sanofi’s alleged common course of conduct.

Filed-rate doctrine

In PNE Energy Supply LLC v Eversource Energy,^[10] the First Circuit addressed a follow-on case to its 2019 decision, Breidling v Eversource Energy,^[11] regarding application of the filed-rate doctrine to claims alleging manipulation of the market for wholesale natural gas sales regulated by the Federal Energy Regulatory Commission (FERC). As in the earlier case, the First Circuit held that the plaintiffs’ claims were barred by the filed-rate doctrine.^[12]

In Breidling, the First Circuit held that antitrust and consumer-protection claims against two energy companies by a putative class of retail electricity customers were barred by the filed-rate doctrine because they impermissibly asked the court to second-guess regulatory judgments made by FERC.^[13] Specifically, the Breidling plaintiffs alleged that the defendants had artificially restricted the supply of natural gas by consistently reserving more pipeline capacity than they required and then declining to resell their excess capacity, which had downstream effects on prices for wholesale electricity and ultimately raised the prices paid by consumers. The First Circuit concluded that such claims were barred by the filed-rate doctrine because it implicated a FERC-approved tariff applicable to interstate pipelines. That tariff allows direct purchasers of natural gas to enter into ‘no notice’ contracts with pipeline operators under which they (1) may adjust their reservations of pipeline capacity upward or downward without incurring penalties and (2) have the right, but not the obligation, to resell their unneeded transportation capacity to other natural gas purchasers.^[14] The First Circuit concluded that the plaintiffs’ claims of market manipulation effectively challenged the ‘no notice’ contracts authorized and regulated by FERC, and thus dismissed the suit as interfering with FERC’s exclusive regulatory authority.

In the new appeal, PNE Energy Supply, the First Circuit addressed a parallel action brought against the same two energy companies by a putative class of wholesaler energy purchasers based on the same basic theory of market manipulation through the reservation of excessive pipeline capacity that the defendants then declined to resell. Not surprisingly, the First Circuit reached the same conclusion in holding that the claims were barred by the filed-rate doctrine since, as the court explained, the status of the plaintiffs as wholesalers or retail purchasers was not relevant to that doctrine’s application.

The First Circuit rejected the plaintiffs’ attempts to distinguish their legal theories from Breidling as involving a refusal to deal in a short-term ‘secondary capacity market,’ and the manipulation of a price index for natural gas through activities that drove up the average price of natural gas. The First Circuit reasoned that these theories merely repackaged the type of claim rejected in Breidling, because they ultimately turned on an allegation that the defendants had driven up prices by declining to release pipeline capacity. Finally, the First Circuit rejected a broader challenge for the court to reconsider application of the filed-rate doctrine to capacity-reserving decisions, since they were not affirmatively approved by FERC but simply allowed by the agency as part of a market-based system of price regulation. The court explained that whatever the policy merit of this critique, it should be directed to FERC or Congress, not federal courts.

Criminal enforcement

In United States v Vega-Martinez,^[15] the First Circuit affirmed the criminal convictions of two school bus operators who engaged in a bid-rigging and market-allocation conspiracy with other bus operators serving a municipal school system. Under the conspiracy, the defendants and others agreed in advance who would submit the ‘low’ bid for different bus routes as part of an auction process administered by the municipality, thus preventing competition. In affirming the convictions, the court held that there was sufficient evidence to satisfy the Sherman Act’s requirement of a sufficient effect on interstate commerce, relying on both the fact that the conspiracy targeted federal education funds and the fact that the defendant’s scheme could have had an impact on the number of buses the municipality would purchase from outside Puerto Rico.^[16]

The court also rejected one of the defendant’s contentions that he should have been able to contend that the process of renegotiating bids after the auction had resulted in reasonable final prices: the court explained that such an argument was precluded by the fact that both bid rigging and market allocation are per se violations of the Sherman Act.^[17]

‘Pay for delay’ litigation

In the Intuniv litigation,^[18] which involves allegations that a brand and a generic drug manufacturer entered into a ‘pay for delay’ patent settlement agreement that delayed generic entry, the district court addressed several motions for summary judgment and related Daubert motions. The court’s primary decisions fall into four categories.

First, the court addressed whether the evidence created a triable issue on the existence of a reverse payment. The plaintiffs alleged that the defendants had agreed that the brand company would not sell an authorized generic version of the brand product (an AG) during the generic company’s regulatory 180-day exclusivity, and courts have found that a no-AG agreement can operate as a reverse payment under FTC v Actavis.^[19] What makes this case unusual is that the settlement agreement expressly reserved the brand company’s right to sell an AG. The plaintiffs alleged that, even so, the economic terms of the agreement – which would have caused the royalty paid by the generic during the 180-day period to drop from 25 percent to zero if an AG launched – meant that the defendants had reached an implied no-AG agreement, on the theory that both parties understood that the economic ramifications of the royalty term would have made it irrational to launch the AG, so the brand had effectively guaranteed to the generic that it would not face an AG. The plaintiffs and defendants offered expert testimony on the settlement negotiations and the economic implications of the royalty term. The defendants sought summary judgment that there was not sufficient evidence from which a jury properly could find an agreement not to launch an AG. The defendants focused on the explicit language of the agreement itself allowing the brand to launch an AG, the course of negotiations in which the generic company repeatedly asked for a no-AG agreement but the brand refused to give one, and the parties’ subsequent actions, including the brand company’s post-agreement analysis of whether to launch an AG or not. The court denied summary judgment, holding that ‘there is a dispute of material fact . . . about whether the parties secretly agreed to enter into a no-AG agreement.’

Second, the court addressed competing cross-motions for summary judgment on market definition and monopoly power. The plaintiffs presented evidence relating to (1) the brand company’s profit margins, (2) the degree to which the generic price was lower than the pre-generic brand price, (3) expansion in the use of the product following generic entry, and (4) the existence of the alleged reverse payment itself, all of which the plaintiffs contended to be direct evidence of monopoly power. The plaintiffs also presented an econometric analysis that purported to show the lack of cross-price elasticity between the brand and other products. Last, the plaintiffs offered the opinion of a physician who stated that the brand product was not interchangeable with other medicines.

In turn, the defendants presented evidence (1) criticizing the purported direct evidence of monopoly power, (2) criticizing the econometric study as relying on data that did not capture net price, and (3) showing competition between the brand and other products at formularies through rebating and the use of restrictions to steer patients to other products, the use of co-pay coupons to compete with generic versions of other products, marketing expenditures, and elasticity studies showing that the brand’s market share against products with similar indications expanded as its prices decreased. Each side sought to exclude the other’s experts, criticized the other’s evidence, and sought summary judgment. The court largely declined to exclude the experts’ testimony and held that both the direct and indirect evidence presented fact disputes about market definition and monopoly power that a jury would have to resolve.

Third, the court addressed the admissibility of expert testimony concerning the merits of the underlying patent litigation between the brand and the generic defendants, an issue that goes to causation of harm in the antitrust case. Although government enforcers need only to prove a substantial likelihood that challenged conduct will lead to anticompetitive effects, private plaintiffs seeking damages must prove that the conduct caused them actual harm.^[20] In this case, the plaintiffs sought to prove that the challenged agreement caused them to overpay because, absent the challenged agreement, lower-priced generics purportedly would have been available earlier. As part of that case, the plaintiffs offered the expert opinion of a law professor that the generic challenger had a greater than 95 percent likelihood of winning the patent case. The defendants challenged the admissibility of that opinion on several grounds, including that the expert did not use any reliable methodology to reach his opinion quantifying the percentage likelihood that the generic would have won. The court ruled that the plaintiffs’ expert would be allowed to testify that, in his opinion, the generic company was more likely than not to prevail in the patent litigation, had it not settled. The court noted in addition that an expert ‘may also be permitted to offer a specific estimate of the likelihood of success in an underlying trial if she relies on a sufficiently testable methodology.’ However, the court found that the expert here had offered no methodology for how he had arrived at his 95 percent figure, so the court excluded any testimony about ‘any specific percentage of likelihood’ to support his opinion that the generic was likely to win.

Fourth, the court addressed the proper application of the ‘Actavis Inference,’ which certain commentators and courts have recognized as standing for the proposition that if the brand made a large and unjustified reverse payment to the generic in connection with settling patent litigation, it is reasonable to infer that the payment may have caused actual delay in generic entry.^[21] The plaintiffs’ economist planned to testify that the Actavis Inference operates as an ‘if/then’ inference, meaning that if there is a reverse payment, then there must have been generic delay (i.e., that the inference is mandatory, not permissive). On that basis, the expert planned to testify that because, in his opinion, there had been a reverse payment, generic entry necessarily must have been delayed. The defendants challenged this opinion as contrary to the First Circuit’s Nexium decision, which affirmed a jury’s finding that even though there had been a large and unexplained reverse payment, that payment caused no actual delay in generic entry. The Intuniv court agreed with the defendants and held that the plaintiffs’ expert ‘shall not be permitted to testify that a large payment necessarily means that there had to be a delayed entry.’

Antitrust counterclaims to patent litigation

Nuance Communications, Inc v Omilia Natural Language Solutions, Ltd involves antitrust counterclaims brought by a defendant in patent litigation.^[22] Nuance and Omilia are competitors in the sale of automated speech recognition technology. Nuance sued Omilia for infringing certain patents. Omilia responded by filing counterclaims alleging improper monopolization in violation of section 2 of the Sherman Act, improper acquisitions of competitors and patents in violation of section 7 of the Clayton Act, and state law violations based on unfair competition and tortious interference. Omilia alleges that Nuance was created out of multiple corporate acquisitions and that it had acquired more than 5,000 patents relating to voice recognition technology. Omilia also alleges that Nuance has a share of more than 70 percent of the relevant market.

At the crux of its antitrust claim, Omilia alleges that Nuance engaged in a strategy of ‘acquir[ing] actual and potential competitors through a calculated scheme of threatening to assert and/or actually asserting baseless patent infringement litigation using its massive portfolio of acquired patents to drive its competitors out of the market and/or coerce them into being acquired by Nuance.’ As part of this scheme, Nuance also allegedly contacted the targets’ customers about the alleged patent infringement, causing the targets to lose commercial opportunities. Omilia alleges that Nuance had employed this scheme against at least 17 competitors, including Omelia. Nuance’s conduct harms competition, Omilia alleges, because it allows Nuance to maintain supracompetitive prices without innovating, which it otherwise would have been forced to do.

Nuance moved to dismiss Omilia’s counterclaims, but the court largely denied that motion. First, Nuance argued that all its alleged conduct was independently lawful and, therefore, cannot constitute monopolizing conduct. The court disagreed, citing authority that otherwise lawful conduct may violate the antitrust laws when undertaken as part of a scheme to monopolize. Nuance also argued that its litigation conduct was protected by Noerr-Pennington immunity, but the court declined to address the applicability of that doctrine on a motion to dismiss. Second, Nuance asserted that Omalia had failed to properly allege a relevant geographic market, because Omilia alleged a market limited to the United States, but competition for speech recognition products occurs worldwide. The court found that this issue could not be resolved on a motion to dismiss, particularly given that much of the alleged conduct involved Nuance’s use of US patents to exclude competitors within the United States. Third, the court rejected Nuance’s arguments that the charges were time-barred, because even if some of the acquisitions at issue had occurred long ago, the effect of those acquisitions on Omilia had ‘ripened into a prohibited effect’ and allegedly caused Omilia injury within the statute of limitations.

Private actions challenging acquisitions

Bio-Rad Laboratories, Inc v 10X Genomics, Inc, also arising from antitrust counterclaims brought by a patent defendant, involved 10X’s challenge to Bio-Rad’s acquisition of a third party.^[23] Bio-Rad and 10X compete along several dimensions in the market for life science tools used for genetic research. Each has several patents and has sued the other for infringement in several cases. In this matter, 10X challenged Bio-Rad’s 2017 acquisition of RainDance, which owned patents for one of the genetic research technologies relating to the parties’ business. 10X alleged that this acquisition constituted an improper acquisition in violation of section 7 of the Clayton Act, and improper actual or attempted monopolization in violation of section 2 of the Sherman Act, in three separate markets. In general, 10X alleged that the RainDance acquisition allowed Bio-Rad to achieve near-monopoly power by eliminating a competitor and consolidating RainDance’s patents with its own. As a result, 10X alleged, Bio-Rad launched increased patent suits against 10X, so 10X faced the costs of defense and the risk of being excluded; 10X was forced to incur increased costs to innovate around Bio-Rad’s patents; and 10X either was unable to obtain licenses from Bio-Rad or had to pay increased licensing fees. 10X sought treble damages and an order requiring Bio-Rad to divest certain patents.

On Bio-Rad’s motion, the court largely dismissed 10X’s claims but allowed two to go forward. First, the court held that 10X’s claims concerning the litigation brought by Bio-Rad failed under the Noerr-Pennington doctrine and could not be used to establish antitrust liability even as a part of a larger pattern of practice. Second, the court refused to allow 10X to proceed based on claims that Bio-Rad would not license 10X, because there is generally no duty to aid a competitor. Similarly, 10X’s arguments about being forced to innovate in light of Bio-Rad’s patents could not support antitrust liability, because that effect is not antitrust injury; on the contrary, it goes to the heart of competition. Third, with respect to two of the three alleged markets, the court found that 10X’s allegations could not plausibly support the required showing of market or monopoly power. With respect to one of the alleged markets, however, the court found that 10X’s allegations plausibly stated both Sherman Act and Clayton Act claims, based on Bio-Rad already having a 90 percent share and then acquiring a competitor, causing 10X to pay supracompetitive royalties that would have been lower absent the acquisition. Even while noting that 10X’s allegations ‘barely pass the threshold of specificity required,’ the court allowed those claims to proceed. The court similarly denied Bio-Rad’s argument to reject the claim for divestiture, on the grounds that the challenged acquisition had occurred fewer than four years prior to the complaint and 10X did not seek a complete unwinding of the deal but only divestiture of certain patents.

Notes