Skip to main content

Second Circuit Holds a Hard Switch Between Drugs Is an Unlawful Product Hop Under Section 2

On May 22, 2015, in a much-watched case, the Second Circuit upheld a preliminary injunction against Actavis PLC and its wholly owned subsidiary, Forest Laboratories, LLC (collectively “Actavis” or “Forest”), finding that Actavis’s “hard switch” strategy to launch an extended-release version of its blockbuster Alzheimer’s therapy and delist the immediate-release version would likely violate Section 2 of the Sherman Act.1 The Court held that because generic competition depends heavily on state drug substitution laws that allow pharmacists to substitute generics for branded products, the combination of launch and product removal constituted an anticompetitive “product hop” that would likely impede generic competition on the merits for the original immediate-release version of the drug.

Factual and Procedural Background

In January 2004, Forest launched its twice-daily immediate-release therapy for moderate-to-severe Alzheimer’s, Namenda IR (memantine)2, generating over $1.5 billion in annual sales for the last several years. Five generic manufacturers have tentative FDA approval to launch on July 11, 2015 when regulatory exclusivity expires, and seven more may be approved for entry as early as October 2015. This generic entry was expected to reduce Forest’s share of the immediate-release memantine market by over 80% within six months.

Forest developed a once-daily, extended-release version, Namenda XR, to improve patient compliance with treatment and extend its Alzheimer’s franchise. Newly issued patents covering Namenda XR prevent generic entry until 2029. FDA approved Namenda XR in June 2010 and Forest launched in July 2013. As part of its initial “soft switch” launch strategy, Forest stopped actively marketing IR; promoted XR to physicians, caregivers, patients, and pharmacists; and discounted XR so patients would pay a lower co-pay for XR than IR. Internal Forest forecasts estimated that this “soft switch” strategy would only net a 30% market conversion.

The Court found in the record that, in order to boost the switch rate to an estimated 80 to 90%, Forest instituted several new strategies, which the Court called a “hard switch”: (1) on February 14, 2014, Forest publically announced it would discontinue Namenda IR in August 2014, (2) Forest notified FDA of its intent to delist Namenda IR, and (3) Forest requested that the Centers for Medicare & Medicaid Services remove IR from its formulary so Medicare plans would not cover its prescription or use.

The State of New York filed its complaint in September 2014, alleging that the planned withdrawal of Namenda IR violated § 2 of the Sherman Act, and seeking a preliminary injunction requiring Forest to keep Namenda IR on the market at least through licensed generic entry in July 2015. The district court granted the preliminary injunction, finding that the State was likely to succeed on the merits and that competition and consumers would suffer irreparable harm if Forest’s proposed actions were not halted.

The district court made several critical findings: (1) withdrawing IR from the market forced Alzheimer’s patients to switch to XR as the only available alternative; (2) generic IR is not equivalent to XR and, thus, is not substitutable3 for branded XR; (3) high transactions costs would make patients forced to switch from IR to XR unlikely to switch back to generic IR; and (4) Forest’s explicit purpose for withdrawing IR from the market was “to impede generic competition.”4 Forest appealed to the Second Circuit.

Second Circuit’s Decision and Analysis

The Second Circuit focused its antitrust analysis on the question of whether Forest’s actions in removing Namenda IR from the market coerced patients and physicians into switching to Namenda XR. Critically, the Court found that the combination of launching a new product and delisting the original product resulted in the anticompetitive effects of sustaining and maintaining Forest’s control over the memantine market. On its own, launching a new product was procompetitive, but in conjunction with removing the old product from the market, that combination could run afoul of Section 2’s prohibition against monopolization.

The Court recognized that “[a]s a general rule, courts are properly skeptical about claims that competition has been harmed by a dominant firm’s product design changes.” United States v. Microsoft, 253 F.3d 34, 58-60 (D.C. Cir. 2001) (en banc). Nonetheless, the Court noted that “[w]ell-established case law makes clear that product redesign is anticompetitive when it coerces consumers and impedes competition.”5 The Second Circuit turned to the analysis in its 35 year-old Berkey Photo, Inc. v. East Kodak Co. case, 603 F.2d 263 (2d Cir. 1979). There, Kodak, with a lawful monopoly in film, but not in cameras, introduced the new Kodacolor II film usable only in a new Kodak camera. The Court rejected a claim that Kodak was unlawfully using its film monopoly to monopolize the camera market, but at the same time (in a footnote) cautioned that “the situation would be completely different if, upon the introduction of the [new system], Kodak had ceased producing film in [old size], thereby compelling camera purchases to buy a [new Kodak camera].” Id. at 287. n. 39.

Applying this analysis here, the Court concluded that the hard switch was coercive because (1) it forced current Namenda IR patients to switch to the new Namenda XR in order to maintain their treatment regimen without interruption, and (2) transactions costs would likely prevent patients from switching back to Namenda IR upon generic entry. The Second Circuit reasoned: “Certainly, neither product withdrawal nor product improvement alone is anticompetitive. But under Berkey Photo, when a monopolist combines product withdrawal with some other conduct, the overall effect of which is to coerce consumers rather than persuade them on the merits, and to impede competition, its actions are anticompetitive under the Sherman Act.”6

The Court rejected the defendants’ arguments that courts should not distinguish between hard and soft switches by again invoking Berkey: “the market can determine whether one product is superior to another only ‘so long as the free choice of consumers is preserved.’”7 Citing Microsoft, the Court found that forcing patients to switch to XR would prevent generic substitution, the hard switch would eliminate what would otherwise have been competition on the merits.8

The Court also rejected Forest’s “free riding” argument that the antitrust laws do not protect competitors simply free riding on the innovations or competitive efforts of others. Forest suggested that generic manufacturers should advertise their products, lobby insurance companies for placement on formularies, and seek prior authorization requirements for their products over other competitors. But, according to the Court, that promotion is not practical or cost-effective because there is no guarantee that a pharmacist would substitute that particular generic for the branded product instead of any other approved generic.9 Critically to the Court, Congress and the state legislatures had made a policy decision to endorse generic “free riding” by developing Hatch-Waxman and the automatic substitution laws.

In addition to the coercion question, the Court recognized the long-held antitrust tenet that a willingness to forsake short-term profits in favor of long-term monopolization or market power was evidence of anticompetitive intent. Here, the Court couched the withdrawal of Namenda IR from the market as a willingness to forego short-term profits – with revenues in excess of $1.5 billion annually since 2012 – in favor of achieving long-term market power by switching the market to Namenda XR. This evidence was persuasive to the Court.

The Second Circuit used the Supreme Court’s recent decision in FTC v. Actavis, Inc., 133 S. Ct. 2223 (2013) to reject Defendants’ argument that their patent rights shield them from antitrust liability. The Defendants argued that they had merely exercised rights afforded by the Patent Act. But to the Court, “it is the combination of Defendants’ withdrawal of IR and introduction of XR in the context of generic substitution laws that places their conduct beyond the scope of their patent right for IR or XR individually.”10

Lastly, the Court upheld the unorthodox remedy that Forest must continue to produce Namenda IR and must not delist the product prior to anticipated generic entry in July because either path would irreparably harm competition and consumers. Allowing the switch in any way would cause “[p]ermanent damage to competition” and economically harm consumers by imposing higher prices to the tune of nearly $300 million for consumers, $1.4 billion for third-party payors, and $6 billion for Medicare over ten years.11 Importantly, these significant sums and potential for irreparable harm to competition by effectuating the switch and effectively excluding all future generic competition were sufficient to impose this highly unusual injunctive relief.


The opinion leaves several critical takeaways. First, this was an issue of first impression with the Courts of Appeal, and it is ironic that it arose in the Second Circuit, which then utilized dicta in its somewhat dated Berkey opinion upon which to ground its analysis. Several other similar cases have reached the same result – that is, when a pharmaceutical company delists its branded product in advance of generic entry, then it is a violation of the antitrust laws. The District Court of Delaware came to this result nine years ago in Abbott Labs. v. Teva Pharm. USA, Inc. when Abbott pulled various dosage strengths of TriCor from the market and the Eastern District of Pennsylvania reached the same result just six months ago regarding Reckitt Benckiser’s Suboxone lifecycle management strategy.12

Second, the Court implicitly endorsed the “soft switch” strategy as inherently procompetitive, suggesting that it was a form of price and product superiority competition that the antitrust laws encourage. According to the Court, the “soft switch” was not coercive and allowed patients and physicians to “evaluate the products and their generics on the merits in furtherance of competitive objectives.”13 The “hard switch,” that is, the removal of Namenda IR from the market, was the only potentially anticompetitive action because it effectively forced patient conversion. This result is in line with the results of other pharmaceutical cases. Nonetheless, it probably overstates the case to suggest a “soft switch” is per se legal – one can anticipate challenges to such programs as having coercive elements that puts it within the holding here, but a well-crafted program should withstand antitrust scrutiny.

Third, the Court considered innovation and the incentives to innovate to be a significant potential procompetitive justification or benefit, but indicated that it will look skeptically upon innovation arguments that appear pretextual.14 Here, the Court found that there was no significant innovation between the immediate-release and extended-release product for at least two reasons: (1) Namenda IR was the sole twice-daily Alzheimer’s treatment on the market and all other drugs were once-daily treatments, like Namenda XR, and (2) branded success here would incentivize incremental innovative changes such as dosage changes or timed-release formulations with minimal risk rather than larger, more significant new chemical developments with greater market risk.15 The Court believed it would embolden larger investment in research and development instead of marginal changes.

Finally, this case is another example of the considerable antitrust product market narrowing in Hatch-Waxman cases over the past few years. Here, the Court found (and, indeed, the parties did not dispute) that the product market was the molecule, rather than the broader market for Alzheimer’s treatments.16 This limited market definition essentially leads a court to skip the market power analysis and nearly assume that the branded pharmaceutical company has market power in the market for the particular molecule. This Court’s endorsement only furthers that oversimplification.




1 State of New York v. Actavis PLC, No. 14-4624 (2nd Cir. May 22, 2015) [hereinafter referred to as Slip Op.].

2 Memantine partially blocks the brain’s n-methyl d-aspartate (“NMDA”) receptor by preventing the receptor from becoming overexcited, which can cause neuron toxicity and brain degeneration. Another class of Alzheimer’s therapies are acetylcholinesterase inhibitors (“CIs”) that work by reducing the breakdown of chemical messengers transmitting information between nerve cells in the brain. Because of these differences, the Court found that the drugs are prescribed in conjunction with, not independently of, memantine, and, as a result, memantine was a relevant antitrust market on its own.

3 State automatic substitution laws, for the most part, require that a generic product be bioequivalent and be AB-rated to their branded alternative in order to be automatically substituted. Because Namenda XR is a different formulation of memantine and at a different dosage level, generic Namenda IR cannot be automatically substituted. As a result, generic companies must file a new Abbreviated New Drug Application seeking FDA approval for the new formulation and dosage level, which is also likely to result in new patent litigation over the patents specifically listed for Namenda XR.

4 Slip Op. at 22-23.

5 Slip Op. at 33.

6 Slip Op. at 35-36 (citations omitted).

7 Slip Op at 38, citing Berkey, 603 F.2d at 287.

8 Slip Op. at 39-40, citing Microsoft, 253 F.3d at 65.

9 Slip Op. at 40-41, 45.

10 Slip Op. at 52-53.

11 Slip Op. at 56.

12 See, e.g., Abbott Labs. v. Teva Pharm. USA, Inc., 432 F. Supp. 2d 408, 424, 430 (D. Del. 2006) (denying the motion to dismiss the antitrust claims because it was likely anticompetitive to remove old versions of TriCor from the market, including buying back stock of the older product, prior to expiration of various regulatory exclusivities); In re Suboxone (Buprenorphine Hydrochloride & Naloxone) Antitrust Litig., 2014 U.S. Dist. LEXIS 167204, *21-32 (E.D. Pa. Dec. 3, 2014) (allowing the antitrust claims to proceed on grounds that the active removal of the old product from the market, as opposed to soft switch tactics, could constitute an antitrust violation).

13 Slip Op. at 37.

14 Slip Op. at 50.

15 Slip Op. at 16.

16 Slip Op. at 30.

Subscribe To Viewpoints


Bruce D. Sokler

Member / Co-chair, Antitrust Practice

Bruce D. Sokler is a Mintz antitrust attorney. His antitrust experience includes litigation, class actions, government merger reviews and investigations, and cartel-related issues. Bruce focuses on the health care, communications, and retail industries, from start-ups to Fortune 100 companies.