Minerva Surgical v. Hologic: Assignor Estoppel Survives, but Only Within Its Fairness Limits

Minerva Surgical, Inc. v. Hologic, Inc., 594 U.S. 559 (2021), asked whether an inventor who sells his patent rights may later turn around and tell a court those rights are worthless. The doctrine that ordinarily forbids that maneuver — assignor estoppel — had been criticized as both a relic and an obstacle to weeding out bad patents. In a 5-4 decision authored by Justice Kagan, the Court refused to discard the doctrine but sharply limited it: assignor estoppel applies only when an assignor’s later invalidity challenge contradicts an explicit or implicit representation he made in assigning the patent. Because that principle can dissolve when the assigned claims are later broadened, the Court vacated the Federal Circuit’s judgment and remanded for a closer look at what the inventor had actually warranted.

At a glance

• Case: Minerva Surgical, Inc. v. Hologic, Inc., 594 U.S. 559 (2021), 141 S. Ct. 2298, Docket No. 20-440
• Court: Supreme Court of the United States, on certiorari to the U.S. Court of Appeals for the Federal Circuit
• Decided: June 29, 2021; 5-4
• Opinion: Justice Kagan, for the Court (joined by Roberts, Breyer, Sotomayor, and Kavanaugh); Justice Alito dissenting; Justice Barrett dissenting (joined by Thomas and Gorsuch)
• Subject matter: Whether assignor estoppel bars a patent’s assignor from challenging the patent’s validity in later infringement litigation
• Holding: Assignor estoppel remains valid but is limited to instances where the assignor’s invalidity claim contradicts explicit or implicit representations made in assigning the patent

The inventor, the assignment, and the broadened claim

The dispute grew out of the career of Csaba Truckai, an inventor of medical devices. Truckai invented a device for treating abnormal uterine bleeding and, with colleagues, founded a company called NovaCept to commercialize it. He assigned his interest in the invention and the related patent applications to NovaCept. Through a chain of corporate acquisitions, those rights eventually came to rest with Hologic, Inc., which continued to prosecute the applications before the Patent Office.

Truckai later left and founded Minerva Surgical, which developed a competing device. Hologic, meanwhile, obtained a continuation patent with a claim broad enough to cover Minerva’s product. When Hologic sued for infringement, Minerva argued the asserted claim was invalid. Hologic invoked assignor estoppel, contending that Truckai — and thus his company, Minerva — could not attack the validity of a patent he had assigned. The district court and the Federal Circuit agreed, applying the doctrine to bar Minerva’s invalidity defense.

The wrinkle that reached the Supreme Court was the breadth of the asserted claim. Minerva contended that the claim Hologic ultimately obtained was broader than anything Truckai had assigned, because Hologic had expanded the claim during continued prosecution to capture Minerva’s later device. If so, Minerva argued, Truckai never warranted the validity of that broadened claim, and estopping him from challenging it made no sense.

Justice Kagan’s middle path: keep the doctrine, cabin it to its rationale

The Court framed the question historically. Assignor estoppel has deep roots — the Court had recognized a version of it in Westinghouse Electric & Manufacturing Co. v. Formica Insulation Co., 266 U.S. 342 (1924), analogizing it to the older rule that a seller of real property cannot later deny that he had good title to convey. The equitable core is fairness: an assignor who represents that a patent is valid, and is paid for that representation, should not be free to take the money and then destroy the very right he sold. “It would be unfair,” the Court explained, to allow an assignor to reap the benefit of the assignment and then contradict himself to the buyer’s detriment.

But the Court refused to treat that fairness principle as boundless. Minerva and the United States as amicus had pressed the Court to abolish assignor estoppel outright, arguing it conflicts with the patent statute’s policy of purging invalid patents and with cases like Lear, Inc. v. Adkins, 395 U.S. 653 (1969), which abolished the parallel doctrine of licensee estoppel. The Court declined. Congress had legislated against the backdrop of assignor estoppel without disturbing it, and the doctrine’s fairness rationale retained force.

The decisive move was to hold that the doctrine reaches only as far as its premise. Assignor estoppel “applies when, but only when, the assignor’s claim of invalidity contradicts explicit or implicit representations he made in assigning the patent.” Where the assignor made no such warranty, the doctrine’s foundation disappears. The Court then offered concrete examples of that mismatch — and one of them fit Minerva’s theory precisely.

When the assignor never warranted the claim

The Court identified three situations in which an assignor’s later invalidity argument does not contradict any prior representation. First, an assignment may occur before any warranty of a specific patent’s validity is even possible — as when an employee assigns rights to all future inventions as a condition of employment, long before any application is drafted. Second, a later change in the governing law may mean that what the assignor once warranted as valid is no longer measured by the same standard, so the assignor’s new position does not contradict the old one.

Third — and central to the remand — the assignee may return to the Patent Office after the assignment and secure claims materially broader than those the assignor assigned. “Assuming that the new claims are materially broader than the old claims,” the Court reasoned, “the assignor did not warrant to the new claims’ validity.” He could not have, because those claims did not exist when he made his representations. To estop him from challenging claims he never warranted would extend the doctrine past the only justification that supports it.

Because the lower courts had applied assignor estoppel categorically without deciding whether Hologic’s asserted claim was in fact broader than what Truckai assigned, the Court vacated and remanded. On remand, the Federal Circuit would have to determine whether the continuation claim covering Minerva’s device was materially broader than the claims Truckai originally assigned — and if it was, assignor estoppel would not bar Minerva’s invalidity defense.

Open questions

Minerva preserved assignor estoppel but left its operation to be worked out case by case. The pivotal new question — when are later claims “materially broader” than those the assignor assigned? — has no bright-line answer, and continuation practice routinely produces claims that differ in scope from the originals. Courts must now compare warranted claims to asserted claims and gauge the mismatch, an inquiry that resembles claim-scope analysis but serves an equitable purpose. The decision also leaves unsettled how implicit representations are to be inferred from different kinds of assignments, and how the doctrine interacts with the strong statutory interest in eliminating invalid patents that animated the dissents. Justice Barrett’s dissent, joined by Thomas and Gorsuch, would have held that the 1952 Patent Act left no room for the judge-made doctrine at all, while Justice Alito’s separate dissent argued the Court could not sensibly reaffirm Westinghouse without deciding whether that precedent should stand.

Implications

• Assignor estoppel is alive but conditional. Counsel can no longer treat an assignment as an automatic bar to every invalidity defense; the defense survives whenever the challenge does not contradict a representation the assignor actually made.
• Broadened continuation claims are vulnerable. An assignee that expands claims after the assignment to capture a former inventor’s new product may forfeit estoppel as to those broadened claims — a direct check on a common competitive strategy.
• Assignment documents now do real work. Drafters should make the scope of an inventor’s warranty explicit, because what the assignor represented — and did not — determines how far estoppel reaches.
• Employee “future invention” assignments carry weak estoppel. Broad, pre-invention assignments of all future work product may not support estoppel at all, since the employee warranted nothing about any specific later patent.

Frequently asked questions

What is assignor estoppel? Assignor estoppel is an equitable doctrine that bars a person who has assigned a patent (typically the inventor) from later arguing, in an infringement suit, that the assigned patent is invalid. It rests on the idea that it would be unfair to let a seller warrant a patent’s value and then attack that value after taking the buyer’s money.

Did Minerva abolish assignor estoppel? No. By a 5-4 vote the Court declined to abolish the doctrine, holding it remains good law. But it narrowed the doctrine to reach only invalidity arguments that contradict what the assignor explicitly or implicitly represented when assigning the patent.

When does assignor estoppel NOT apply after Minerva? The Court identified situations where the assignor made no relevant warranty of validity: for example, an employee’s broad assignment of future inventions, a change in governing law after the assignment, or, importantly, where the assignee later amends the claims to be materially broader than those the assignor assigned. In that last situation the assignor never warranted the new, broader claims.

Authorities and sources

• Minerva Surgical, Inc. v. Hologic, Inc., 594 U.S. 559 (2021), 141 S. Ct. 2298, Docket No. 20-440 (decided June 29, 2021). Justia; official slip opinion (Supreme Court).
• Oral argument and case summary via Oyez.
• Vote (5-4), Kagan authorship, and the Alito and Barrett dissents corroborated by Wikipedia: Minerva Surgical, Inc. v. Hologic, Inc..
• Historical roots of the doctrine in Westinghouse Elec. & Mfg. Co. v. Formica Insulation Co., 266 U.S. 342 (1924), via Justia.