University Inventions, Stanford vs. Roche

Can a university “retain” something it never had to begin with?  Can the federal government “receive” rights from inventors with whom it has no contractual relationship?  In post-Bayh-Dole America, with whom do rights to innovation first vest?

These are some of the questions the United States Supreme Court may have been interested in addressing when it recently granted Stanford University’s petition for a writ of certiorari to review the decision of the United States Court of Appeals for the Federal Circuit in Board of Trustees of the Leland Stanford Jr. University v. Roche Molecular Systems, Inc., 583 F.3d 832 (Fed. Cir. 2009), cert. granted, No. 09-1159 (U.S. 2010).  The case raises the question of who owns the patent rights to a set of methods for quantifying the amount of Human Immunodeficiency Virus (“HIV”) in human blood samples and correlating those measurements to the therapeutic effectiveness of certain antiretroviral drugs.  The technology was invented by several researchers at Stanford University, including Dr. Mark Holodniy, as part of a program in which Stanford, a federally funded university, collaborated with Cetus Corporation, a start-up research company.  The ruling below turned on whether the University is sole owner of the patents on these inventions.  Relying upon the University and Small Business Patent Procedures Act of 1980 (35 U.S.C. § 200 et seq.), commonly known as the Bayh-Dole Act, the Court of Appeals ruled that the University owns the patents only in part and thus lacks standing to sue for infringement.

Human Inventors, Federally Funded Universities, and Private Start-Ups

The complexity surrounding ownership of these inventions arose in part from competing intellectual property agreements that Dr. Holodniy signed in order to work for Stanford in collaboration with Cetus.  When he joined Stanford in 1988, he signed a form agreement with Stanford that recited: “I agree to assign . . . to Stanford” the inventions developed under Stanford’s contracts with the federal government.  When Stanford sent Holodniy to visit Cetus in 1989, he signed another form agreement with Cetus that recited: “I will assign and do hereby assign to CETUS” the rights to the inventions developed as a consequence of his access to Cetus’s technology.  In 1995, Holodniy executed an assignment of his rights in the inventions to Stanford.  But meanwhile Roche, who had purchased Cetus and Cetus’s agreements with Stanford and its researchers, had begun manufacturing HIV detection kits using the Holodniy technology.  See generally 583 F.3d at 837-38, 841-42.

Stanford sued Roche for patent infringement in 2005.  Roche claimed that Stanford lacked standing to sue because Roche, through its purchase of Cetus, was itself an owner of Holodniy’s interest in the asserted patents.  One of Stanford’s responses was that, under the Bayh-Dole Act, Stanford had a prior right to Holodniy’s ownership interest in the inventions because they arose in connection with federal funding, and therefore Holodniy had no present ownership interest to assign to Cetus when he purported to transfer it.  The district court agreed.  The Federal Circuit reversed, and held that Holodniy had an ownership interest in the inventions in 1989 and had transferred that interest to Cetus upon signing the Cetus agreement.  According to the Federal Circuit, this left nothing to transfer to Stanford, and nothing in the Bayh-Dole Act altered this result.

Stanford framed the question presented to the Supreme Court as:

Whether a federal contractor university’s statutory right under the Bayh-Dole Act in inventions arising from federally funded research can be terminated unilaterally by an individual inventor through a separate agreement purporting to assign the inventor’s rights to a third party.

Roche framed the question differently:

Whether the Boyh-Dole Act permits a private university to void, unilaterally and retroactively, an inventor’s otherwise valid contractual assignment of his rights in an invention, solely because that invention was developed in part with federal funds.

In its grant of certiorari, the Supreme Court accepted Stanford’s framing of the question.

The Federal Circuit’s Analysis

The Federal Circuit’s analysis required a journey through numerous bodies of law, both state and federal, common and statutory.  In order to see how the Bayh-Dole Act fit into that analysis, it is helpful to break out and in some cases re-order the steps in the Federal Circuit’s analysis.

Although unstated in Stanford v. Roche, the Federal Circuit finds the origin of rights in an invention in the common law: “the act of invention itself vests an inventor with a common law or ‘natural’ right to make, use and sell his or her invention.”  Arachnid, Inc. v. Merit Indus., Inc., 939 F.2d 1574, 1578 (Fed. Cir. 1991).  The Federal Circuit skipped immediately to the discussion of the effects of the various contracts that Holodniy signed on his rights in his inventions, but the next step is to consider whether the Bayh-Dole Act altered this original, “natural” vesting of the inventor’s rights in his own invention.  The Federal Circuit held that it did not: “Nothing in the [Bayh-Dole Act], regulations, or our caselaw indicates that title is automatically forfeited.”  Stanford, 583 F.3d at 844 (quoting Central Admixture Pharm. Servs. Inc. v. Advanced Cardiac Solutions, P.C., 482 F.3d 1347, 1352-53 (Fed. Cir. 2007)).  At this point, then, title remained with Holodniy.

The chain of title would next depend on the contractual assignments of Holidniy’s rights in his inventions.  The Federal Circuit explained that the question of who owns a patent is ordinarily a question of state law, but that patent assignment clauses are so patent-specific as to warrant their own body of federal contract law.  See Stanford, 583 F.3d at 841.  Under Federal Circuit law, Holodniy’s contract with Stanford, which stated “I agree to assign,” was merely a promise to assign rights in the future, so nothing was transferred when he signed the contract in 1988.  However, Holodniy’s contract with Cetus stated “I will assign and do hereby assign,” and therefore effected a present assignment of Holodniy’s future inventions to Cetus immediately upon signing in 1989.  Thus, Holodniy transferred all of his rights to Cetus in 1989, and Stanford’s purported “purchase” of these rights in 1995 was not effective because after 1989, Holodniy had nothing left to sell.  Id. at 841-844.

Having followed the chain of title through its contractual transfers, the Federal Circuit then considered again whether the Bayh-Dole Act had any subsequent effects on the ownership of the Holodniy interest in the inventions.  The court held that the Act gave the government the right to take title to Holodniy’s ownership interest, and in turn to void his assignment of that interest to Cetus, but that the government did not exercise those rights.  Id. at 844.  The court held that the Bayh-Dole Act gave Stanford the right to elect to “retain ownership if the Government does not,” but that since Holodniy had already alienated his rights in the invention in 1989, and the government had not voided those rights and taken title itself, the rights remained where they lay (in Cetus) and Stanford had nothing to “retain.”  Id.  The Bayh-Dole Act did not “automatically void ab initio the inventors’ rights in government-funded inventions,” nor did it “void[ Holodniy’s] prior contractual transfers of rights.”  Id.

Questions For The Supreme Court

There are two things conspicuously absent from the Federal Circuit’s analysis, which the Supreme Court may address when it renders its decision.

The first is a close look at the relevant statutory text.  As the Supreme Court reaffirmed last term, even in patent law, the starting point for a statutory analysis should be the words of the statute, and the Court will ordinarily interpret these words in accordance with “dictionary definitions,” “common usage,” and “ordinary meaning.”  Bilski v. Kappos, 130 S. Ct. 3218, 3225-26 (2010).  The Bayh-Dole Act states that a federally funded contractor such as a university “may . . . elect to retain title to any subject invention.”  35 U.S.C. § 202(a) (emphasis added).  Merriam-Webster’s online dictionary lists several definitions of “retain,” but, in most of them, “retain” does not mean to acquire ownership but rather to maintain ownership over something already in possession:

  • to keep in possession or use
  • to keep in mind or memory
  • to hold secure or intact

The one exception is “to employ by paying a retainer,” for example, “to retain an attorney,” but this is hardly the ordinary meaning or common usage of the word; it is hard to imagine that Stanford elected to “retain” a patent ownership interest the way one retains an attorney.  Thus, the Act permits a contractor to “retain” title to inventions that it owns—a reversal of the previous legal landscape, in which ownership often automatically vested in the government—but the Act is silent as to where title originates or how it gets to the possession of the contractor to begin with.

Stanford and some amici have argued that this interpretation would defeat the purpose of the Bayh-Dole Act, which is to facilitate the exploitation of federally funded research by non-government entities.  But even if the Supreme Court agrees that the law’s silence as to the origination of title and manner by which such title passes to a contractor allows for frustration of the act’s purpose (which it might not), the Court may be unwilling to re-write the statute in a way that ties up these loose ends.

Also conspicuously absent from the Federal Circuit’s decision is an explanation for its implicit holding that the Act empowers the government to void otherwise valid transfers of ownership but not a government contractor.  The Act states that the government “may receive title to any subject invention in which the contractor does not elect to retain rights or fails to elect rights within” the prescribed time, 35 U.S.C. § 202(c)(2) (emphasis added), and “may receive title to any subject invention” that the contractor does not disclose to the government within a reasonable time, 35 U.S.C. § 202(c)(1) (emphasis added).  In a series of opinions, the Federal Circuit has held that although this provision does not automatically void ab initio the inventor’s rights in government-funded inventions, it allows the government to “take title” and “invoke forfeiture” of the contractor’s title, such that the contractor’s “title to the patent may be voidable.”  Stanford, 583 F.3d at 844; Central Admixture, 482 F.3d at 1352-53, Campbell Plastics Engineering & Mfg., Inc. v. Brown, 389 F.3d 1243, 1249-50 (Fed. Cir. 2004).

In Central Admixture, however, the government did not attempt to exercise its right to void the inventor’s title to the subject invention, and the Federal Circuit “assuredly [did] not decide” whether it had that right.  482 F.3d at 1353.  And in Campbell Plastics, although the government did exercise its right to void its contractor’s title to the subject invention, there was no dispute that the contractor itself held these rights.  389 F.3d at 1246.  These were not cases in which the government attempted to take title from the contractor but the contractor had no rights for the government to take because the contractor’s employee had already assigned his rights to a private company.  Had the government in Stanford decided it did want to “receive” title to the subject inventions in 1995 after Stanford’s ineffective election to “retain” what it never had, would the Federal Circuit have held that the government had the power to void the title transfer of Holodniy, who was not himself a contractor, to Cetus, who was also not a contractor?  If so, this would mean that the government can do what Stanford could not.  But what, if anything in the Act creates this distinction?  While “receive” and “retain” are not the same word, and thus may be understood to create differences between the government’s ability to obtain title and the contractor’s, neither “receive” nor “retain” carries the same meaning or connotation as “take,” “void,” or “invoke forfeiture.”

In sum, by granting certiorari in Stanford v. Roche, the Supreme Court has indicated a willingness to look closely at the interplay of patent ownership rights under competing common law and statutory frameworks.  In the process, the Court may very well address the Federal Circuit’s statutory interpretations as announced not only in Stanford but in its earlier Bayh-Dole cases.

By Mark A. Kressel, Irell & Manella LLP, Los Angeles

All images and illustrations used in our posts are licensed and have been legally acquired through official sources such as Adobe Stock