The story of Cephalon and its blockbuster sleep-disorder drug Provigil is the cleanest illustration in modern pharmaceutical antitrust of how two distinct strategies — reverse-payment settlements and product hopping — can be combined to extend a patent monopoly years beyond what the patent itself would support. This Article uses the Provigil case to map the mechanics of that combination and to argue that the harm to consumers and to the integrity of the Hatch-Waxman framework is greater than either strategy considered alone.
I. Introduction
Provigil (modafinil) was approved by the Food and Drug Administration in 1998 to treat narcolepsy and other sleep disorders. By the mid-2000s it had grown into a billion-dollar product for Cephalon, the small Pennsylvania-based pharmaceutical company that held the U.S. rights. Cephalon’s composition-of-matter patent on modafinil expired in 2001, and the only patent left protecting Provigil was a narrow formulation patent that the company itself acknowledged was vulnerable to challenge. Four generic manufacturers filed Abbreviated New Drug Applications under the Hatch-Waxman Act in 2002 and 2003, certifying that they would not infringe the formulation patent. The stage was set for an ordinary Hatch-Waxman dispute that should have ended, within a few years, with generic competition for Provigil.
It did not. By 2012, when generic Provigil finally reached the market, Cephalon had already shifted most of its prescribers and patients to a successor product, Nuvigil, that enjoyed patent protection until 2023. The Federal Trade Commission later estimated that the delay had cost American consumers and insurers more than four billion dollars. The story of how Cephalon achieved this result deserves close attention because it shows how two strategies that look defensible in isolation produce, in combination, a result that the Hatch-Waxman Act was specifically designed to prevent.
II. Provigil
The narrow formulation patent on Provigil, U.S. Patent No. RE37,516, claimed a particular crystalline form of modafinil. The generic challengers argued that they could manufacture modafinil in a non-infringing form and that the patent itself was invalid because the claimed crystalline form had been on sale more than a year before the patent application. Internal Cephalon documents introduced in later FTC litigation suggest that Cephalon’s own scientists agreed: the patent was, in the company’s candid internal assessment, a weak one.
Faced with that assessment, Cephalon could have tried the patent in court and accepted the result. Instead, in 2005 and 2006, the company entered into settlement agreements with all four generic challengers. The settlements followed the pattern that has come to be known as the reverse-payment or pay-for-delay model: Cephalon paid the generics, collectively, more than $200 million in cash and side deals, and the generics agreed not to enter the market until April 2012, six years before the formulation patent would expire and four to five years later than they would have entered if even one of them had prevailed in court.
III. The Reverse-Payment Settlement
The defense of reverse-payment settlements has always rested on a single comparison. Proponents observe that the settlement permits generic entry years before the formal expiration of the patent. They invite the reader to compare the settled date with patent expiration and conclude that consumers are better off than they would be under no settlement at all. Cephalon made exactly this argument in its motion to dismiss the FTC complaint, noting that the settlement “resulted in generic entry years earlier than patent expiration” in 2015.
The comparison is the wrong one. The right comparison is between the settled date and the date on which the generics would have entered absent the settlement. That date depends on the strength of the patent. For a strong patent the two comparisons converge. For a weak patent — which Cephalon’s formulation patent was — the absent-settlement date is the date the generics would have prevailed in court, which on the available evidence was substantially earlier than April 2012. The settlement therefore did not accelerate generic entry. It delayed it, by purchasing the generics’ agreement not to attempt the challenge that would otherwise have ended Cephalon’s monopoly.
IV. Product Hopping
The reverse-payment settlement was the first half of Cephalon’s strategy. The second half was product hopping. Cephalon had spent years developing Nuvigil, a follow-on product whose active ingredient was armodafinil, the longer-lived enantiomer of modafinil. The Food and Drug Administration approved Nuvigil in 2007. Nuvigil’s composition-of-matter patent ran to 2023.
Product hopping requires two operations: making the original product less attractive and making the successor more attractive. Cephalon performed both. Between 2004 and 2008 the company increased the price of Provigil by approximately 74 percent. In 2009 it ceased substantially all promotion of Provigil; the chief executive later acknowledged in an earnings call that the sales force had “pulled all promotion from Provigil” in anticipation of the Nuvigil launch. Simultaneously, Cephalon mobilized its full sales infrastructure behind Nuvigil, deployed its representatives to physicians, and offered Nuvigil at a modest discount to the artificially inflated price of Provigil.
The result was predictable. Insurers and pharmacy benefit managers, observing that Nuvigil was now cheaper than Provigil and was actively promoted by the manufacturer, began to substitute Nuvigil prescriptions for Provigil prescriptions. By the time the generic versions of Provigil arrived in 2012, the prescribing patterns of treating physicians had largely moved to Nuvigil, and the generic Provigil market that the Hatch-Waxman Act was supposed to create was a fraction of what it would have been in 2007 or 2008.
V. Combination of Settlements and Product Hopping
Each strategy considered alone has its critics, but each has also found defenders. The reverse-payment settlement, defenders argue, is a routine settlement of patent litigation; the parties bargain over the date of generic entry just as they would bargain over any other litigation outcome. Product hopping, defenders argue, is ordinary product evolution; pharmaceutical companies have always invested in successor drugs, and the market should reward those investments.
The Provigil case shows why the combination is qualitatively different from either component. The settlement bought Cephalon a guaranteed window of monopoly. The product hop converted that window into a permanent advantage by moving the customer base to a drug whose patent ran to 2023. Neither move alone would have produced the result Cephalon achieved. Without the settlement, generic competition would have eroded Provigil’s pricing power and made the inflated-price stage of the product hop impossible. Without the product hop, generic Provigil in 2012 would have captured most of the market and the settlement would have been a delay rather than a transformation.
The interaction between the two strategies is what makes the Provigil case important. Antitrust analysis of either strategy in isolation will systematically understate the harm that the combination produces. The work of scholars studying indirect IP enforcement has shown how patent strategies can produce cumulative effects that look small step by step and large in aggregate. The Provigil case is a particularly clean example of the same phenomenon in pharmaceutical markets.
VI. Conclusion
In 2006 Cephalon faced imminent generic entry on Provigil. The generics had a strong claim that they did not infringe Cephalon’s narrow formulation patent. There was no guarantee that the FDA would approve Nuvigil before the generics arrived. Cephalon’s response was to pay the generics not to enter the market and, in the years it bought, to switch the market to Nuvigil. The result was four billion dollars in sales that, in the words of one Cephalon executive, “no one expected.”
The Hatch-Waxman Act was designed to encourage patent challenges by giving the first generic challenger a 180-day exclusivity period. The Provigil case shows how a sufficiently coordinated combination of settlements and product evolution can defeat that incentive. Antitrust enforcement and Hatch-Waxman doctrine should be willing to look at the combined effect of these strategies, not just at the strategies one at a time. As the journal’s analysis of Hatch-Waxman pharmaceutical patents has documented, the statutory framework remains the central battleground for these disputes, and its effectiveness depends on whether enforcers see the full picture.
Frequently Asked Questions
What is a reverse-payment settlement?
A reverse-payment settlement is a deal in which a brand-name pharmaceutical company pays a generic manufacturer to delay entry into the market. The ‘reverse’ refers to the unusual direction of the payment: in ordinary patent litigation the alleged infringer pays the patent holder, but in these settlements the patent holder pays the alleged infringer to drop its challenge. The Federal Trade Commission has long argued that these arrangements harm consumers by extending the patent holder’s monopoly.
What is product hopping?
Product hopping is a strategy by which a brand-name drug company shifts its market from one formulation to a successor formulation, typically before generic versions of the original become available. By making the original less attractive and the successor more attractive, the company can transfer its customer base to the successor and minimize the damage done by eventual generic entry.
How did Cephalon use both strategies with Provigil?
Cephalon paid four generic manufacturers more than $200 million collectively to delay their generic versions of Provigil until 2012. During the years bought by these settlements, Cephalon raised the price of Provigil by 74 percent, ended its promotion, and launched and aggressively marketed Nuvigil, a successor drug protected by a patent running until 2023. By the time generic Provigil entered the market, the customer base had largely moved to Nuvigil.
Related articles: Hatch-Waxman Pharmaceutical Patents · IP Privateering · Bilski and Life Science Patents · The Medicines Patent Pool · Patent Prosecution Highway