For many companies that compete in competitive markets, innovation and product improvement fuels competitive success: create a better product than your competitors, and over time, make it better. This holds true in the pharmaceutical industry, where new and/or improved treatments are common and frequent for almost any indication. In recent years however, certain product modifications, marketed as improvements, have given rise to antitrust challenges.
“Product hopping” refers to the pharmaceutical practice of a branded pharmaceutical company switching patients from an older version of a drug, typically for which patent expiration and generic competition are imminent, to a new version of the same drug with some modification and no immediate threat of generic competition. The product switch may hamper competition, especially if the older product is removed from the market, because generic drug manufacturers depend on automatic substitution to divert prescriptions from the branded product to the generic version of that product. The Federal Trade Commission (“FTC”), in 2012, first stated that product hopping “can be an effective way to game the regulatory structure that governs the approval and sale of generic drugs, thereby frustrating the efforts of federal and state policymakers to facilitate price competition in pharmaceutical markets.”1
Antitrust challenges to product hopping are rare, but recently two different appellate courts issued significant opinions in such cases.2 Both of these cases have implications for the antitrust treatment of product hopping allegations. The most recent case, decided in part on the issue of product market definition, may have broader antitrust implications as well.
Pharmaceutical Industry Background
The pharmaceutical industry has unique complexities. Generally, Food and Drug Administration (“FDA”) approval and patent protection limits the number of products available for any particular indication. To be approved for sale, companies must a submit a New Drug Application (“NDA”), and they are required to, among other items, submit scientific studies and conduct clinical trials, which are costly and time-consuming processes. The Drug Price Competition and Patent Term Restoration Act (the “Hatch-Waxman Act”) counters these limitations by creating a truncated approval process for generics known as the Abbreviated New Drug Application (“ANDA”).3 This process allows generic drug companies to rely on a name brand drug company’s approved NDA for a particular drug.4 “By enabling generic manufacturers to ‘piggyback on a brand drug’s scientific studies’ and the significant costs associated with their NDA, Hatch-Waxman ‘speeds the introduction of low-cost generic drugs to market, thereby furthering drug competition.’”5 To rely on a name-brand’s NDA, however, the generic drug manufacturer must demonstrate that the proposed generic product is both equivalent to the name-brand drug in active ingredients, dosage form, strength, and route of administration (pharmaceutical equivalence) and yields similar test results for human absorption (bioequivalence).6 Approved generics are deemed therapeutically equivalent or “AB-rated” to the corresponding branded drug (also known as a Reference Listed Drug, or “RLD”).
AB-rated generic drugs have a major market advantage through state automatic substitution laws. These laws permit or require pharmacists to substitute an AB-rated, lower-cost generic drug for a brand drug absent express direction from the prescribing physician.7 Due to lower upfront costs, generic drugs are generally less expensive than branded drugs and usually capture a substantial share of the prescriptions that would otherwise go to the brand. Generics drugs typically prompt a revenue and profit loss to the brand company that faces an onset of generic competition.
A product hop prevents generic drugs from benefiting from the state substitution laws. ANDAs to a branded drug generally are not substitutable to an AB-rated equivalent to a different formulation of the branded drug, even if the difference is insignificant.8 A brand’s decision to change an RLD’s dosage/strength (e.g., 10 mg tablets versus 20 mg), the form of the drug (e.g., tablet versus capsule), or the absorption qualities of the drug (immediate versus extended release), can all make a drug a different product such that a generic version of the old drug would not be AB-rated for the new one. These changes may give the branded drug fresh patent protection against generic competition. So long as the old version of the drug remains on the market, a doctor may still prescribe that version, allowing its approved AB-rated generics to benefit from automatic substitution. If, however, the brand company successfully removes the old version of the drug from the market before generic entry occurs, then doctors must switch the patients to other products or the newer RLD and will likely no longer prescribe the old version of the drug once a generic of that version reaches the market. In such circumstances, the product switch may completely remove the imminent threat of generic competition.
State of New York v. Actavis plc
In State of New York v. Actavis plc (“Namenda”), the Second Circuit became the first appellate court to address product hopping.9 The State of New York alleged that Forest Laboratories (“Forest”), a subsidiary of Actavis plc, planned to remove its twice-daily Alzheimer’s drug, Namenda IR, from the market, and thereby force all patients to its new once-daily drug, Namenda XR with patent exclusivity through 2029. Both drugs contain the same active ingredient memantine. New York alleged that this was a coercive switch designed to avoid generic competition and monopolize the memantine market.10 Forest had patent exclusivity preventing the marketing of generic Namenda IR until July 2015, after which Forest would likely lose most of its sales to generics within six months. This period is referred to as the “patent cliff.”11 In 2013, XR was introduced as a “soft switch,” which means Forest aggressively marketed Namenda XR to doctors, patients, and pharmacists, and sold XR at a discount that made it less expensive than IR. In addition, Forest, effectively lowered the price of XR through rebates to health plans, aimed at limiting co-payments for the drug. Subsequently, in early 2014, when Forest faced imminent generic competition, Forest commenced a “hard switch” from IR to XR. The company announced in February 2014, that they would discontinue Namenda IR in August 2014, and provided notice to the FDA and the Centers for Medicare and Medicaid Services of the discontinuation. Production issues caused a delay, at which time New York sued to enjoin the discontinuation, prompting Forest to agree to a standstill in September 2014.12
In addition to the allegations above, New York alleged that XR offered no material benefit to patients over IR to justify the switch. Finally, the parties did not dispute that the relevant product market was memantine, for which Namenda IR and XR were the only products on the market at the relevant time. The court held a five-day evidentiary hearing and found: (1) Withdrawing Namenda IR from the market prior to generic entry leaves Namenda XR as the only available alternative for Alzheimer’s patients dependent on memantine therapy; (2) Generic versions of IR poised to enter the market in 2015 will not be AB-rated to XR; (3) Pharmacists will not be permitted to automatically substitute generic IR for Namenda XR because IR and XR are not therapeutically equivalent; (4) If Forest forced patients to switch to XR prior to generic entry, those patients would likely not switch back to twice-daily IR (for generic) due to high transaction costs associated with patients switching formulations; (5) Preventing generic IR from competing under state drug substitution laws would effectively prevent generic entry into the memantine drug market; and (6) Forest’s “explicit purpose” for the switch was to impede generic competition through state substitution laws.13 The district court concluded that New York raised serious questions under Sections 1 and 2 of the Sherman Act (and the state law equivalent), showed a risk of irreparable harm, and that the balance of equities favored an injunction. The court granted New York’s request and issued a preliminary injunction requiring that Forest continue to manufacture and supply Namenda IR (on the 2013 terms), notify the proper channels of IR availability, and avoid other measures that would inhibit generic substation of IR. The injunction carried until July 2015, 30 days after generic IR was slated to reach the market.14
Affirming the lower court, the Second Circuit found that Forest held a lawful monopoly on Namenda IR, but that that the product hop at issue was anticompetitive and exclusionary conduct aimed at achieving an unlawful continuation of that monopoly.15 The court deemed the hard switch an act of “coercion”:
• Defendants’ hard switch crosses the line from persuasion to coercion and is anticompetitive. As long as Defendants sought to persuade patients and their doctors to switch from Namenda IR to Namenda XR while both were on the market (the soft switch) and with generic IR drugs on the horizon, patients and doctors could evaluate the products and their generics on the merits in furtherance of competitive objectives. By effectively withdrawing Namenda IR prior to generic entry, Defendants forced patients to switch from Namenda IR to XR―the only other memantine drug on the market.16
The court stated that “[w]hether XR is superior to IR is not significant in this case,” given Defendant’s clear intention to avoid the patent cliff and the coercion that removed consumer preference as a market force.17 The court further found Forest’s purported justification and procompetitive arguments for the hard switch to be “pretextual,” or otherwise impermissible, including the argument that generic drug companies seek to free-ride on the demand created by branded drugs, and product hopping benefits competition by blocking free-riding.18 The court also rejected the notion that Forest was maximizing their return on investment, finding, “‘the willingness to forsake short-term profits to achieve an anticompetitive end’ is indicative of anticompetitive behavior,” and observing that Forest would not recoup $1.5 billion in annual IR sales absent the benefit of reduced competition.19 The Second Circuit held that the hard switch constituted monopolization and attempted monopolization, and did not consider the Section 1 or state law claims. The Court affirmed the preliminary injunction.
Mylan Pharmaceuticals, Inc. v. Warner Chilcott PLC
More recently, in Mylan Pharmaceuticals Inc. v. Warner Chilcott PLC, the Third Circuit addressed product hopping allegations concerning Doryx, an oral antibiotic of the tetracycline class used to treat severe acne.20 In 2005, Defendants Warner Chilcott and Mayne introduced a new tablet form of Doryx to replace the incumbent capsule version of the product, for which Plaintiff Mylan attempted an AB-rated generic.21 The capsule version of the drug was discontinued for sale, and removed from Warner Chilcott’s website. Warner Chilcott also bought back and destroyed Doryx capsules that were already on the market. Later, in 2007, 2008 and 2010, the Defendants sought, and ultimately gained in the following year, approval of subsequent tablet versions of the drugs, each time changing the dosages and scores. For example, instead of a 75 mg tablet, Warner Chilcott manufactured a scored 150 mg tablet that patients could split in half, and, later, a dual-scored tablet that could be split in thirds. These modifications would require a generic to file a new ANDA demonstrating the bioequivalence and pharmaceutical equivalence between their drug and the reformulated branded drug for an AB-rated generic. In each case, Warner Chilcott later discontinued the prior versions of the drug. Mylan sought FDA approval of an AB-rated generic for each tablet version that Warner Chilcott introduced, and alleged that Warner Chilcott’s product hopping, with exceptions, prevented Mylan from obtaining FDA approval of a generic version of Doryx that would benefit from automatic substitution.22 Mylan obtained ANDA approval for the 2008 version of Doryx tablets in February 2012, by which time Warner Chilcott was already marketing the dual-scored 2010 version. Mylan then sued in July 2012, alleging that Warner Chilcott monopolized the Doryx market through a continuous pattern of product hopping.
Mylan alleged that Defendants violated Sections 1 and 2 of the Sherman Act, in addition to state law claims. Mylan specifically alleged that the four new versions of Doryx did not offer therapeutic benefits or improve the drug in any material way, but were rather mechanisms to prevent generics from reaping the benefits of automatic substitution and to frustrate their efforts to market a generic version of Doryx.23
The Eastern District of Pennsylvania granted summary judgment for the Defendants. The court agreed that Warner Chilcott may have engaged in product hopping to delay market entry, but found that each of Mylan’s antitrust claims failed.24 Specifically, the court found that the relevant product market was oral tetracyclines prescribed to treat acne, rather than Doryx, specifically. In that market, the Doryx market share was a mere 18%, well below the threshold at which courts may find monopoly power.25 The district court further found that market was characterized by freedom of entry and little product differentiation.26 The court held that Warner Chilcott lacked monopoly power in this broader market. Warner Chilcott therefore could not be held liable for monopolization under Section 2 of the Sherman Act. The court also found that Defendants’ conduct did not prevent competition for Doryx sales. Mylan was free to (and did) introduce AB-rated generics of each version of Doryx, which doctors were, in turn, free to prescribe. The court found that Mylan made no effort to market the generic drugs, instead relying only on automatic substitution (which is how generic drugs reach the market generally). The court expressed concern that finding a loss of automatic substitution to be an anticompetitive injury would “trigger . . . a burden-shifting contest” in which any product modification would subject a branded company to litigation, which would chill drug innovation.27 Finding no harm to competition, the court held there was no Sherman Act violation as a matter of law.
On appeal, the Third Circuit affirmed. The court first held that Warner Chilcott did not have monopoly power in a relevant product market. The court affirmed the district court holding that Mylan did not provide sufficient direct evidence of monopoly power based on supracompetitive prices or restricted output, or specifically “price-cost margins.” The court found that Mylan’s deficiency was a failure of proof through expert testimony, rather than any particular facts favorable to Warner-Chilcott.
Finding insufficient direct evidence of monopoly power, the Third Circuit turned to indirect evidence of monopoly power. Key to this finding is the relevant product market definition. Consistent with many cases addressing brand pharmaceutical conduct, Mylan alleged a market consisting of Doryx and its generic equivalents. The court assessed whether other products were interchangeable with Doryx and demonstrated cross-elasticity of demand with Doryx.28 On interchangeability, the court focused on dermatologist consensus, FDA labeling, health insurance/formulary treatment, and managed care organization reimbursements, all of which indicated that other tetracyclines “were fully substitutable for Doryx.”29 Evaluating demand, the court found “the undisputed evidence demonstrates that ‘when Defendants increased the price of Doryx, its sales decreased and the sales of other oral tetracyclines increased,’”30 and further credited expert testimony that Warner-Chilcott’s sales and marketing strategy was both motivated by and had a material impact on other tetracyclines. Based on a “high degree of interchangeability and cross-elasticity demonstrated in the record,” the Third Circuit agreed with the district court that Doryx competes in a broader market for tetracycline oral antibiotics to treat severe acne, and rejected Mylan’s argument of a Doryx-specific submarket, finding Doryx “is interchangeable with a wide variety of other tetracyclines,” and not recognized as a “separate economic entity.”31 In this broader market, Warner Chilcott’s market share never exceeded 18%, and thus there was no indirect evidence of monopoly power.32
The Third Circuit then applied the rule of reason framework set forth in United States v. Microsoft, which looks at whether a defendant’s conduct is of an anticompetitive nature and further whether defendant can show “‘nonpretextual procompetitive justifications for its conduct.’”33 With respect to the first prong, the court agreed with the district court that Mylan was not entirely blocked from the market, noting Mylan’s sales of generic Doryx during the relevant time period, and that the original Doryx capsules were off patent and faced other generic competition for many years prior to Mylan’s ANDA.34 The court distinguished Namenda in two important respects.35 First, the timing and conduct in Namenda was driven by pending patent expiration, prior to which no generic could access the market. Second, Forest attempted and intended to remove Namenda IR from the market prior to a generic launch, which the court found to constitute coercion (a forced switch). Neither of these facts were present in the case of Doryx. The court also found that Warner Chilcott “offered strong evidence of non-pretextual purposes for their various product changes,” including difficulties related to capsule consumption, shelf-life stability and competitive response.36 The Third Circuit found that Warner Chilcott did not have monopoly power and that their conduct was not anticompetitive. The panel therefore affirmed that there was no violation of Sections 1 or 2 of the Sherman Act.
Mylan’s request for rehearing en banc was recently denied by the Third Circuit.37 An amicus brief filed by the FTC in support of rehearing en banc sought clarification on two general issues in the opinion.38 The FTC raised concern that the appellate opinion limited the means by which Mylan could prove monopoly power. Specifically, the agency suggested that monopoly power may be shown “through direct evidence of detrimental effects,” such as “eliminating or limiting generic competition through product hopping or other exclusionary conduct.”39 The FTC argued that the steep price declines of brand drugs upon generic entry support the argument that the brand and generic comprise their own relevant product market, and that a brand is unlikely to benefit from product hopping unless it already had monopoly power to protect.40 The other issue raised by the FTC was that the appellate panel erred in its analysis of exclusionary conduct on two fronts. First, the opinion focused on the impact of product hopping on Mylan rather than the overall impact on competition, which is the proper nucleus for analyzing competitive effects. Second, the FTC claimed that the panel “misunderstood the degree of foreclosure necessary to prove exclusionary conduct,” by addressing whether Mylan could compete at all, as opposed to whether Mylan was forced into less efficient distribution channels or higher costs of doing business.41 The FTC was specifically concerned that Warner Chilcott may have denied Mylan the benefit of automatic substitution, thereby thwarting Mylan’s ability to compete effectively, notwithstanding the fact that Mylan was able to access the market.
The appellate decisions with respect to Namenda and Doryx, provide appellate authority on the two elements needed to prove a monopolization claim for product hopping. In Namenda, Forest’s monopoly power was undisputed. The Second Circuit never addressed the monopoly power test that led the Pennsylvania court and the Third Circuit to find that Doryx lacked monopoly power. As monopoly power was undisputed in Namenda, the Second Circuit was able to address the alleged conduct, the purported justification, and the competitive effects, ultimately, finding a violation. Namenda is generally a plaintiff-friendly decision and will potentially serve as a valuable precedent for plaintiffs in bringing product hopping suits. But the court’s factual findings in Namenda were critical to its outcome:
• Product Market/Monopoly power – the branded drug and generic equivalent were in the same submarket
• Imminent patent cliff – the alleged product hopping was tied to the imminent threat of generic entry/penetration following patent expiration
• Absence of non-pretextual benefits or valid business justifications for the product switch
• Removal of the old drug from the market prior to generic entry constituted coercion
In Mylan, the Third Circuit found in the Defendants’ favor and heavily distinguished the circumstances from Namenda. As it currently stands, Mylan could potentially be helpful for defending against product hopping suits—although also limited by its facts. The Second and Third Circuit decisions are potentially valuable opinions for plaintiffs and defendants, respectively, but they are not in conflict. There is little to suggest that the Namenda facts would not have prevailed in the Third Circuit (that court distinguished Namenda), and there is likewise little in the Second Circuit opinion suggesting the Mylan facts would have stated a claim under the Second Circuit test. The Second Circuit specifically found:
• Had Defendants allowed Namenda IR to remain available until generic entry, doctors and Alzheimer’s patients could have decided whether the benefits of switching to once-daily Namenda XR would outweigh the benefits of adhering to twice-daily therapy using less-expensive generic IR (or perhaps lower-priced Namenda IR). By removing Namenda IR from the market prior to generic IR entry, Defendants sought to deprive consumers of that choice. In this way, Defendants could avoid competing against lower-cost generics based on the merits of their redesigned drug by forcing Alzheimer’s patients to take XR, with the knowledge that transaction costs would make the reverse commute by patients from XR to generic IR highly unlikely.42
Both cases found in favor of the party (New York, Warner Chilcott) that prevailed on all four bulleted issues listed above.
The Third Circuit’s finding with respect to monopoly power, and the FTC’s response, have implications far broader than product hopping suits. A finding of monopoly power is essential in any rule of reason case, and whether markets are confined to a single brand product or a broader category has a major impact in pharmaceutical cases alleging reverse payment patent settlements, sham patent litigation, sham citizen petitions, and other suits alleging exclusionary conduct or unreasonable restraints of trade. The FTC, states and private plaintiffs routinely define and allege monopoly power within a market that consists of a branded product and its generics. Competing brands are not considered part of the same market, even if they are reasonable substitutes and/or exhibit cross-elasticity of demand. As discussed in the FTC’s recent amicus brief, the support for the narrow brand-specific market definition derives from the patent cliff, automatic substitution and the price impact of generics relative to any competing brands.43 Decades of post Hatch-Waxman brand conduct instruct the view that brand pharmaceutical owners are uniquely concerned and resistant to potential generic competition and severely impacted by generic penetration, even when faced with inter-brand competition.44
The challenge for the FTC and private plaintiffs, for all branded pharmaceutical conduct cases, is perhaps testing the reach of the Third Circuit’s product market analysis. The indirect evidence relied on by the Third Circuit was thorough, and may exist in other cases where brand pharmaceutical companies are accused of thwarting generic competition. Can a brand’s robust competition in a broader market (e.g., therapeutic class) overcome the presumption that a branded drug and its generic equivalents comprise a valid market? This is a landmark question. Modern pharmaceutical antitrust law suggests the answer is no; pharmaceutical conduct cases alleging anticompetitive conduct targeted at generics typically define the market around the brand regardless of non-brand competition. This is the FTC’s position and a critical point for plaintiffs generally.45 However, the Mylan opinion finds the relevant product market to be the broader market, which could be a significant precedent.
The Third Circuit’s denial of Mylan’s rehearing en banc petition is significant. It will be interesting to see if relevant product market definition becomes a viable argument for pharmaceutical defendants accused of disrupting generic competition, in product hopping suits or more broadly. For the near future, parties should expect that the existence of brand competition is still unlikely to defeat monopoly power with respect to brand-generic allegations, but Mylan offers another clear reminder that fundamental antitrust law like market definition is important, worthy of economic scrutiny, and should be aggressively disputed by branded pharmaceutical defendants where available.
With respect to product hopping, the Namenda and Mylan decisions provide a framework to analyze product hopping, albeit a fragile one. Warner Chilcott’s conduct did not rise to the level that the Second Circuit found to be an antitrust violation and Forest’s conduct would have been a violation under the Third Circuit’s standard. It is unclear how the Second Circuit (or others) will treat product hopping in the absence of a patent cliff and how other courts will handle disputes with respect to monopoly power. On this issue, some courts outside of the Second and Third Circuits may align closer to the FTC position. Finally, the Second and Third Circuits may have shown reluctance to find liability based on soft switches, but the FTC and private plaintiffs will likely test the limits of this theory where the facts suggest coercion.
Product hopping is an emerging area of antitrust law and one worth watching. Pharmaceutical companies should exercise caution and should consult with antitrust counsel when engaging in product improvements that have the potential to impede generic competition.