The 2007 MedImmune decision is significant because the United States Supreme Court in this decision construed and changed the scope of Article III of the United States Constitution. The Court construed Art. III more broadly to make it easier for patent licensees to attack patent validity in federal courts. In particular, the Court applied an old test in a new way to hold that Declaratory Judgment jurisdiction exists for contract disputes even when no party has breached. This is a marked change to prior license law, where an actual breach was needed to support Art. III jurisdiction. The change to Art. III applies both inside and outside the patent world, reaching all corners of the land. Understanding the ramifications of this landmark decision has been challenging, but there are now Federal Circuit decisions applying the new standard in different fact situations. MedImmune and its progeny significantly impact patent license strategies.
MedImmune Inc. v. Genentech Inc.
549 US 118, 127 SCt 764, 81 USPQ2d 1225 (2007)
1. Introduction
The 2007 MedImmune decision is significant because the United States Supreme Court in this decision construed and changed the scope of Article III of the United States Constitution. The Court construed Art. III more broadly to make it easier for patent licensees to attack patent validity in federal courts. In particular, the Court applied an old test in a new way to hold that Declaratory Judgment jurisdiction exists for contract disputes even when no party has breached. This is a marked change to prior licensing law, where an actual breach was needed to support Art. III jurisdiction. The change to Art. III applies both inside and outside the patent world, reaching all corners of the land. Understanding the ramifications of this landmark decision has been challenging, but there are now Federal Circuit decisions applying the new standard in different fact situations.
MedImmune and its progeny significantly impact patent license strategies.
Many patent licenses are placed as a result of arms length negotiation or as a vehicle to settle a dispute. Unfortunately, placing a license is not always a happy ending. Too often a license catalyzes further disputes between the licensor and licensee. The licensee, in particular, becomes disgruntled when it finds the grant clauses are too restrictive or the royalty rate is too burdensome. The disgruntled licensee then seeks ways to avoid its burdens without sacrificing its ability to practice under the licensed intellectual property.
How does a licensee retain rights to practice under patent rights without having to pay for those rights? As one option, the licensee launches a validity attack to nullify the patent rights. A licensee can attack patent validity in one or more different ways. Examples include a U.S. federal court action brought under the Declaratory Judgment (DJ) Act, 28 USC 2201(a) as authorized by Lear v. Adkins, 395 U.S. 653 (1969); reexamination;
post-issue submissions; derivation proceeding; post-grant review; covered business method patent review; inter partes; review; international opposition; inventorship/implied license/waiver; re-negotiation; Walker Process/Antitrust/misuse claims, and the like.
MedImmune involved a validity attack brought in federal court under the DJ Act. The case tested the constitutional limits of federal subject matter jurisdiction. Art. III of the
U.S. Constitution specifies that federal courts may only adjudicate “cases” and “controversies.” Until MedImmune, Art. III required an actual breach of a patent license to support DJ jurisdiction. Without an actual breach, a licensee could not launch a DJ action in federal court to challenge validity. Practically, this forced the licensee to breach its license and put the license at risk to access federal court. If the licensee won and
invalidated the patents, its goals would be achieved. If the licensee lost, licensee is an infringer, has no license, and is out of business. The stakes are high.
From the licensee perspective, a licensee dreams of being able to attack patent validity without being forced to breach its license to gain access to federal court. The MedImmune licensee pursued this dream. Ignoring substantial precedent against the strategy, the non-breaching licensee in MedImmune launched a validity attack in federal court under the DJ Act. The licensor challenged the strategy, arguing that the licensee had no right to access federal court without an actual breach.
Is an actual breach of a contract, in this case a patent license, required to satisfy Art. III of the U.S. Constitution and thereby support the DJ action? And, if an actual breach is not required to satisfy Art. III, then what test or factors are used to assess if Art. III is satisfied in the absence of an actual breach? This is what MedImmune is about.
Constitutional law impacts patent licenses more frequently than might be expected. Brulotte v. Thys 379 U.S. 29 (1964) involved the Discoveries Clause, Art. I, §8, which authorizes Congress to create a patent system in which patents protect inventions for only a limited time. Lear v. Adkins, supra, and its progeny involve separation of powers issues (distinguishing Article I and III tribunals when applying Lear principles to construe blocking clauses) as well as establishing a hierarchy within the Constitution itself (dispute settlement under Article III takes priority over the Discoveries clause). AronsonQuick Point Pencil Co., 440 U.S. 257 (1979) involved a conflict between federal and state powers under Erie v. Tompkins principles.
MedImmune is the most direct clash between licensing law and the U.S. Constitution. The meaning of Article III of the Constitution itself is altered as a result of this Supreme Court decision. Federal courts now have more power but more burdens, too. One can argue that the decision regulates conduct of parties, too, and thereby legislates in violation of separation of powers principles as between the judicial and legislative branches.
Patent license law happens to be the battleground where this Article III battle was fought, but the ramifications affect any contract relationships. The case could provide precedent for the Supreme Court to “legislate” other conduct as well.
2. MedImmune Background.
In 1997 the MedImmune parties entered a patent license relating to a drug used to treat respiratory disease in infants and children. The license encompassed a patent application that issued some years later. In 2001, licensor sent a letter to licensee advising that the newly issued patent covered the drug and that the licensee would have to start paying royalty beginning in March 2002.
Licensee viewed the letter as a threat that licensor would enforce the patent, terminate the license and sue for patent infringement if licensee did not pay the royalty. If licensee lost that infringement action, it could be ordered to pay treble damages and attorney fees and be enjoined from selling the drug. The drug was 80% of licensee’s revenue since 1999. Licensee was unwilling to risk its license by breaching. It paid the royalty, and hence was not in breach, but then brought a DJ action in federal district court to attack the validity of the patent.
Licensor moved to dismiss the DJ action for lack of subject matter jurisdiction. Licensor argued that Art. III only allows federal courts (Art. III tribunals) to adjudicate disputes when there is a case or controversy, and there is no case or controversy in the absence of a breach. Alternatively stated, licensor argued that the DJ Act is unconstitutional if construed to allow the non-breaching licensee’s claim to be heard in federal court.
The dismissal appeared to be a slam dunk under existing law. Guided by the Supreme Court’s legal framework (including the Maryland Casualty case, discussed below) existing until MedImmune, the existing Federal Circuit law provided a clear, definitive framework for analysis with an equally clear, definitive answer. Under Gen-Probe Inc. v. Vysis, Inc., 359 F.3d 1376 (Fed. Cir. 2004), DJ jurisdiction is proper only if there is
The Gen-Probe test was referred to as the “reasonable apprehension of suit” test. Gen-probe held that a patent licensee in good standing cannot establish an Article III case or controversy because licensee cannot possibly have a “reasonable apprehension” that it will be sued for infringement. In the absence of an actual breach, a licensee does not have a ticket to federal court.
The law could not have been clearer. The District Court dismissed the DJ action relying upon Gen-Probe. The Federal Circuit affirmed the dismissal, also relying upon Gen-Probe. As it turned out, both the District Court and the Federal Circuit got it wrong in a tough outcome to predict. The Supreme Court turned the tables on appeal, changing the law, so that licensee and not licensor won the jurisdictional battle.
3. Supreme Court Analysis
The Supreme Court analyzed whether licensee alleged a contract dispute under the DJ Act sufficient to satisfy Article III and thereby avoid dismissal of its claims. Article III, Section 2 of the U.S. Constitution defines the judicial power of federal courts:
The judicial power shall extend to all cases, in law and equity, arising under this Constitution, the laws of the United States, and treaties made, or which shall be made, under their authority;--to all cases affecting ambassadors, other public ministers and consuls;--to all cases of admiralty and maritime jurisdiction;--to controversies to which the United States shall be a party;--to controversies between two or more states;--between a state and citizens of another state;--between citizens of different states;--between citizens of the same state claiming lands under grants of different states, and between a state, or the citizens thereof, and foreign states, citizens or subjects.
The Court began with an analysis of the DJ Act itself. The Court recognized that early on it had doubts whether Article III could encompass declaratory actions generally, citing its precedent from the 1920 time-frame. Not long after that in 1933, though, the Court officially recognized that Article III could encompass declaratory actions in appropriate circumstances. Then the Court specifically upheld the constitutionality of the new DJ Act in 1937 on grounds that the “actual controversy” language of the Act referred to cases and controversies under Article III. Aetna Life Ins. Co. v. Haworth, 300 U.S. 227, 57 S.Ct. 461, 81 L.Ed. 617 (1937). In 1941, the Supreme Court clarified that an “actual controversy” exists when there is a substantial controversy between adverse parties of sufficient immediacy to warrant relief. Maryland Casualty Co. v. Pacific Coal & Oil Co., 312 U.S. 270, 273, 61 S.Ct. 510, 85 L.Ed. 826 (1941). The Maryland Casualty test examines many factors including whether the controversy is definite and concrete, real and substantial, the requested relief is specific and conclusive, the decision is not merely advice or hypothetical, the parties have adverse legal interests, the controversy is sufficiently immediate and real.
The Supreme Court viewed the Maryland Casualty test as controlling, but the court recognized it applied the test to new territory. Had the MedImmune licensee stopped paying royalty, there would be no doubt that Art. III would be satisfied. But, the licensee was in good standing as it continued to make its royalty payments. Assuming that licensor had no claim for an anticipatory breach, licensee had no imminent threat of being sued for infringement as long as it continued to make its payments. The Court framed the issue before it as whether the lack of any immediate threat of being sued causes the dispute to be outside the legitimate scope of Art. III.
Maryland Casualty did not provide the answer. Accordingly, the Court examined other precedent involving criminal (government enforcement) and private actions, respectively, to see if jurisdiction was supported or precluded. The Court observed that there were many Supreme Court cases on point from the criminal field. In those, a party can initiate federal court action to eliminate criminal threats without having to violate the law first. Thus, “[t]he plaintiffs own action (or inaction) in failing to violate the law eliminates the imminent threat of prosecution, but nonetheless does not eliminate Article III jurisdiction.” MedImmune at 772. The criminal cases supported jurisdiction in the absence of a breach.
Criminal cases involve circumstances in which life, liberty and property are at risk to government enforcement. These are not concerns in civil cases like MedImmune. Criminal cases alone could not support where the Supreme Court wanted to take its analysis. Consequently, the Court next analyzed civil cases, which were more sparse.
The Court cited a handful of district court cases that accepted jurisdiction in contract disputes where there was no breach. The Court also mustered two Supreme Court cases that, according to the Court, were on point.
First, the Court cited Altvater v. Freeman, 319 U.S. 359, 63 S.Ct. 1115, 87 L.Ed. 1450 (1943), as supporting jurisdiction when a licensee continues to pay its royalties. Atvater is not so close as the Court makes it appear. The Court twisted and contorted that case so it appears to be directly on point, but that case involved an actual breach of a license. Even if the licensee was still paying its royalties, other parts of the agreement clearly were breached.
The Court attempted to distinguish Willing v. Chicago Auditorium Assn, 277 U.S. 274, 48 S.Ct. 507, 72 L.Ed. 880 (1928). That case refused Article III jurisdiction in the absence of a lease breach in a real estate dispute. The Court disregarded that case as being decided before the DJ Act was enacted. This is not persuasive, of course. It is not seen how enactment of a statute can cause the meaning of the Constitution to change. The statute must comply with the Constitution, not the other way around!
In the end the Supreme Court in an 8-1 decision held that a license breach is not required to support Article III jurisdiction. This is contrasted to Gen-Probe, where an actual breach was required. Interestingly, the Supreme Court in MedImmune and the Federal Circuit in Gen-Probe used the exact same Maryland Casualty (1941) test but each reached contrary outcomes. MedImmune construes the venerable old Maryland Casualty test in a new way. The Supreme Court retired the “reasonable apprehension of suit” test from Gen-Probe without any fanfare. Given the manner by which the Court handled and, in practical effect, ignored decades of its own precedent, the Court pursued this outcome with single minded determination.
Next the Supreme Court applied the Maryland Casualty factors to the facts at hand and held that licensee’s DJ claim satisfied Art. III. As licensee and licensor, the parties had adverse interests. The licensor sent the licensor a letter advising of patent coverage and that the licensee would have to start paying royalties. The Court viewed this as a threat/coercion/affirmative act by the licensor. The threat was immediate. The dispute was significant, as the drug at issue produced 80% of licensee’s annual revenue since 1999. Adjudication resolves the dispute.
MedImmune makes it easier but not effortless for a licensee to launch a validity attack in federal court under the DJ Act. Key factors appear to be an overt act by the licensor, immediacy, and significance.
4. MedImmunizations.
There is no doubt that MedImmune altered the licensing landscape enough that both licensors and licensees need to understand how this case impacts license practice. Each party should consider drafting and negotiation strategies (“MedImmunizations”) to minimize or maximize the impact of the case. Generally, licensors want to incorporate MedImmunizations into their drafting and negotiating while licensees would like their MedImmune rights to be unencumbered. Commentators have suggested several provisions that can be drafted and negotiated to make it harder or easier to launch validity attacks against patents in federal court. The following is a list of candidate strategies, although there is not much case law to guide whether the following strategies pose an antitrust/misuse risk either under a per se or rule of reason analysis:
5. Recent Developments:
5.1 Royalty refunds:
Can a non-breaching licensee get a refund of paid royalty if the licensee attacks validity successfully? The fact scenario involves, at Time A, the licensee initiates a DJ action to attack validity but keeps paying royalty. At Time B, the licensee wins. Can the licensee get a refund of the royalty it paid between Time A and Time B? The Supreme Court expressly refused to answer this question and the answer remains open.
5.2 License Negotiation:
Surprisingly, license negotations trigger DJ jurisdiction even if no overt threats of litigation are made. Sony Electronics Inc. v. Guardian Media Technologies Ltd., 83 USPQ2d 1798 (Fed. Cir. 2007); SanDisk Corp. v. STMicroelectronics Inc., 82 USPQ2d 1173 (Fed. Cir. 2007).
5.3 Settlement Agreements and blocking clauses
Lear v. Adkins revoked licensee estoppel and authorized licensees to challenge validity. MedImmune extended Lear by allowing validity challenges even if the licensee is not in breach. Generally but not always, license clauses that block validity challenges (“blocking clauses”) are not enforceable under Lear or MedImmune. Settlement agreements are an exception in that blocking clauses are enforceable if incorporated into agreements that settle disputes. This includes consent decrees, licenses, covenants not to sue, and the like.
Blocking clauses are enforceable only if the language clearly blocks validity/enforceability attacks. Ambiguous language fails to support blocking. The Federal Circuit demonstrates a bias to construe blocking clauses narrowly. It is also clear that licensees favor challenging settlement agreements. Careful drafting is imperative.
An issue in many cases, therefore, is whether a blocking clause in a settlement agreement actually blocks validity/enforceability challenges as intended. The following are cases where settlement language was clear enough to block validity attacks by the licensee: Hemstreet v. Spiegel, Inc., 7 USPQ2d 1502 (Fed. Cir. 1988); Flex-Foot, Inc. v. CRP, Inc., 57 USPQ2d 1635 (Fed. Cir. 2001); Foster v. Hallco Mfg. Co., Inc., 20 USPQ2d 1241 (Fed. Cir. 1991). These three cases were decided prior to MedImmune. The last two were cited with approval after MedImmune, indicating these cases are still good law.
The following blocking clause is derived from the successful Flex-Foot blocking clause:
Licensee shall not during the term of this agreement or thereafter challenge or cause to be challenged, directly or indirectly, the validity or enforceability of any Licensed Patent in any court or other tribunal anywhere in the world. Licensee waives any and all invalidity and unenforceability defenses in any future litigation, arbitration, administrative proceeding, mediation, or any other proceeding. This clause shall apply to any product or component thereof, composition or component thereof, or method or portion thereof that is made, used, sold, offered for sale, imported, or otherwise transferred by or for Licensee or any assignees, successors or those who act in concert with any of these parties at any time during the life of the Licensed Patents
The following are cases where settlement language was too ambiguous or too narrow to block validity attacks: Baseload Energy Inc. v. Roberts, 96 USPQ2d 1521 (Fed. Cir. 2010); Ecolab, Inc. v.Paraclipse, Inc., 62 USPQ2d 1349 (Fed. Cir. 2002); Diversey Lever, Inc. v. Ecolab, Inc., 52 USPQ2d 1062 (Fed. Cir. 1999); Howmedica Osteonics Corp. v. Wright Medical Technology, Inc., 88 USPQ2d 1129 (Fed. Cir. 2008). All of these cases have been cited with approval after MedImmune, indicating their analyses are still good law.
This language failed to block validity attack (Ecolab. v. Paraclipse): [T]he ‘690 patent is a valid patent.
The following language quite surprisingly failed to block a validity attack (Baseload), because the agreement structure otherwise evidenced an intent not to release patent claims and defenses:
The parties mutually forever release and discharge the other parties . . . From ANY AND ALL losses, liabilities, claims, expenses, demands, and causes of action of EVERY KIND AND NATURE, known and unknown, suspected or unsuspected, disclosed or undisclosed, fixed or contingent, whether direct or indirectly that the parties ever had, now have, or hereafter may have or be able to assert against the other parties . . .
In summary, blocking clauses must be carefully drafted to be effective and there are many counter-intuitive pitfalls. But there are many cases to guide drafting efforts. Flex-Foot, in particular, offers a helping hand and provides an excellent, litigation-tested clause that worked.
5.4 Covenants not to sue (CN2S):
Covenants not to sue (“CN2S”) are a licensor exit strategy to get out of an unwanted DJ action brought by a licensee. Typically, the licensee secured DJ jurisdiction and is attacking validity/infringement. The licensor files/offers up a CN2S along with a motion to dismiss. The licensor argues that the CN2S eliminates a case and controversy so DJ jurisdiction is extinguished. This fact pattern is addressed in many cases including Dow Jones & Co. v. Ablaise Ltd., 95 USPQ2d 1366 (Fed. Cir. 2010); Revolution Eyewear Inc. Aspex Eyewear Inc., 89 USPQ2d 1885 (Fed. Cir. 2009); Benitec Australia, Ltd. V. Nucleonics, Inc. 83 USPQ2d 1449 (Fed. Cir. 2007); Fort James Corp. v. Solo Cup Co., 75 USPQ2d 1257 (Fed. Cir. 2005); Intell. Prop. Dev., Inc. . TCI Cablevision of Calif., Inc., 58 USPQ2d 1681 (Fed. Cir. 2001); Amana Refrig., Inc. v. Quadlux, Inc. 50 USPQ2d 1304 (Fed. Cir. 1999); Super Sack Mfg. Corp. v. Chase Packaging Corp. 35 USPQ2d 1139 (Fed. Cir. 1995).
These cases establish that a timely CN2S of appropriate scope extinguishes DJ jurisdiction. The battles over CN2S documents, not surprisingly, erupt over whether the document is timely and/or of appropriate scope. To be of the right scope, the document must encompass current, past, and specific (non-speculative) future products at issue; past and future infringement; and all the “accused” parties who are to be released.
A CN2S covered only past infringement of current products. The CN2S did not cover past products that were pulled from the market due to litigation. Nor did the CN2S cover any future acts for any products. This CN2S was too narrow to negate a case or controversy under Article III, and the DJ action continued. Revolution Eyewear Inc. v. Aspex Eyewear Inc., 89 USPQ2d 1885 (Fed. Cir. 2009).
A CN2S covered past and future infringement of all products at issue. The CN2S covered subsidiaries, but not the parent. No subsidiaries or parents were parties. This CN2S was sufficient to negate Article III jurisdiction and the DJ action was dismissed. Dow Jones & Co. v. Ablaise Ltd., 95 USPQ2d 1366 (Fed. Cir. 2010).
A CN2S was filed after a jury verdict if invalidity. The Federal Circuit determined that this CN2S was too late to be timely. Fort James Corp. v. Solo Cup Co., 75 USPQ2d 1257 (Fed. Cir. 2005). Amazingly, the Federal Circuit had to reverse the district court!
In summary, a CN2S must be carefully drafted to be timely and have the right scope, but there are many cases to guide drafting efforts.