On February 28, 2011, the U.S. Supreme Court will hear argument in a dispute between Stanford University and the pharmaceutical giant Roche about the effects of an inadvertently-granted patent assignment. It is only the latest of several recent cases involving the validity, scope and effect of patent assignments. A common thread among them is that all arose from the inartful drafting and/or inattention to the details of assignment documents. In other words, they involve disputes that might have been avoided if those involved sweated the details. Here are three very current examples:
1. The legal question now before the U.S. Supreme Court in Bd. of Trustees of the Leland Stanford Junior Univ. v. Roche Molecular Systems, Inc., is whether an apparently inadvertent assignment by a Stanford researcher to a predecessor of Roche of his future inventions trumped both Stanford’s rights, under a prior agreement where the inventor agreed to assign his inventions to the university, and the federal government’s claims to those inventions under the Bayh-Dole Act. That 1980 statute allocates rights between the government and certain private entities in inventions created in federally-funded research.
In the Stanford dispute, the inventor, a Dr. Holodniy, signed a standard agreement with Stanford in which he “agreed to assign” inventions created under federal contracts. Yet, when he visited a company that had done prior work in the field (Cetus, later acquired by Roche), he signed a standard “Visitors Confidentiality Agreement” that, among confidentiality and other provisions, stated that he “hereby assign[ed] to Cetus any right” in inventions conceived “as a consequence of [his] access to Cetus’s facilities or information.” Roche invoked the Visitors Confidentiality Agreement when Stanford sued Roche for infringing inventions Holodniy later created at Stanford.
The Federal Circuit held that the Visitors Agreement was a binding present assignment of after-developed inventions, while Stanford’s agreement was a contract to execute assignments in the future. As a result, Stanford lacked standing to sue Roche for infringement of a patent that Roche in fact owned.
Stanford is now contending that the Federal Circuit’s holding undermines its and the federal government’s rights in inventions, under the Bayh-Dole Act, in the inventions which were discovered in federally-funded research. That statute gives certain private entities conducting government-sponsored research the option of owning inventions they create, and assures the government of at least licensed rights. However the Supreme Court decides that question, however, it is clear that the dispute never would have arisen if either: (1) Stanford’s standard employee agreement contained a present assignment of present and future inventions (that is, “hereby assigns” language), or (2) Holodniy had read, appreciated and objected to the broad language of the Cetus form.
2. In Abraxis Bioscience, Inc. v. Navinta, LLC, 96 USPQ2d 1977 (Fed. Cir. 2010), the Federal Circuit held that Abraxis lacked standing to sue Navinta for infringement of a patent Abraxis acquired (or thought it acquired) in a purchase of an affiliate of Astra-Zeneca. The asset purchase agreement provided that the seller “shall or shall cause one or more of its Affiliates to, Transfer . . . all of the right, title and interest” in the patents of the company being acquired. Because Astra-Zeneca needed to resolve a defect in the chain of title for the patents among its affiliates, however, the patents were not in fact assigned to Abraxis until after it brought suit against Naventa. The Federal Circuit held that: (1) the asset purchase agreement contained only an agreement to assign, not a present assignment, and; (2) the nunc pro tunc assignment to Abraxis after the suit commenced was not effective because, to have standing, a plaintiff must actually own the patents sued upon when it commences litigation.
The dissent in Abraxis noted that Navinta did not seek interlocutory review of the district court’s earlier ruling that Abraxis did own the patents-in-suit, and stressed the inequity of the appellate court’s reversal on this issue after a full infringement trial and three years of litigation. He reasoned the district court properly applied New York contract law, as opposed to the appellate court’s application of federal law.
Whatever the “correct” result might be, Abraxis might have avoided considerable expense and disappointment had it: (1) insisted that chain-of-title problems be resolved before executing the asset purchase agreement; (2) required present assignment language in the purchase agreement, and also included language vesting in it the right to sue on the patents involved; (3) not treated the assignments as unimportant post-closing details, and/or; (4) ensured that it had unassailable title to the patents before commencing litigation.
3. In SIRF Technology, Inc. v. Int’l Trade Comm., 94 USPQ 2d 1607 (Fed. Cir. 2010), SIRF argued that Global Technology lacked standing to challenge SIRF’s importation of allegedly infringing products because, even though the inventor had assigned the patent-in-suit to Global Technology, he had previously granted to his former employer an automatic assignment of present and future inventions “which are related to or useful in the business of [the former employer] . . . and which were . . . conceived during the period of the [employee’s] employment . . . .” It was agreed that the inventor conceived the invention while working for the former employer. Because the invention was also related to the former employer’s business, SIRF alleged, the invention was vested in the former employer, not Global Technology, even though it was reduced to practice and patented while the inventor was at Global Technology.
The Federal Circuit affirmed the ITC’s ruling that the invention was not “related to or useful in” the business of the former employer, and, as a result, Global Technology owned the patent-in-suit and had standing. To do so, it had to look to extrinsic evidence to divine the meaning of the parties, in light of the broad and ambiguous wording of the assignment agreement. It was able to do so because the record of prior, unrelated litigation between the former employer and Global Technology showed that the former employer became aware of Global Technology’s claim to the technology, yet raised no objection and entered into a settlement agreement not addressing the technology.
Global Technology was lucky to have such good evidence of another company’s intent. The decision is a reminder to employers and employees alike that it is in everyone’s interest to define as precisely as possible the scope of assignments (and licenses, non-compete agreements, nonsolicitation agreements, etc.). Where the scope of future business is uncertain, it should nonetheless be possible to define criteria for determining objective boundaries in the future (e.g., products actually sold; budgeted research and development projects). And employers should be reminded to determine the existence and specifics of any prior agreements to which a new hire may be bound.
Assignments often are viewed as standard forms and details to be addressed after the closing of a transaction. Not so. The decisions summarized above, and others, should remind attorneys and business people alike that there is no standard form; that agreements contain language that has implications, sometimes not readily-apparent; that one should not casually sign a document without appreciating its full meaning, and; that untended loose ends can be dangerous. –Christopher J. Mugel
