Open Questions Following Minerva v. Hologic

On June 29, the Supreme Court issued a 5-4 decision in Minerva Surgical, Inc. v. Hologic, Inc., which reaffirmed but limited the patent-law doctrine of assignor estoppel — i.e., estoppel against a patent owner who assigns his rights to another.

Generally speaking, assignor estoppel is an equitable doctrine that prohibits the assignor of a patent (or patent application) from later challenging the validity of the patent (or the claims stemming from the assigned application).

The question in Minerva was whether to uphold the doctrine, narrow it, or get rid of it altogether. The court opted for the middle ground: it declined to discard the doctrine but held that it “applies when, but only when, the assignor’s claim of invalidity contradicts explicit or implicit representations he made in assigning the patent.”

This represents a significantly narrower formulation of the doctrine than the one previously applied by the Federal Circuit.

What does this holding mean in English? It’s probably easiest to explain via a few hypotheticals (adapted from examples provided in the court’s opinion):

1. Sarah sells her issued patent to Wilson. Assignor estoppel bars Sarah from arguing that the claims in the issued patent are invalid because such an argument would contradict her implicit representation that she was selling Wilson something of value.
2. Bill gets a job at Acme Corporation and signs an employment agreement saying that he agrees to assign Acme any patent rights in inventions that Bill develops in the course of his employment. Assignor estoppel does not bar Bill from later arguing that one of those future Acme patents is invalid because Bill could not have made any representations about a patent and invention that do not yet exist.
3. Zac sells his issued patent to Joe in 2021. In 2022, the Supreme Court issues a landmark decision called RSK v. Feletex that changes the law of obviousness. Under the new legal standard, the patent is invalid as obvious. Assignor estoppel does not bar Zac from raising an obviousness challenge to the patent based on RSK, because Zac’s sale carried with it no implicit representation about the validity of the patent in light of a future change in law.
4. Davis sells the rights in his patent application to Nancy. Nancy then files a continuation application that has claims that are materially broader than the claims that Davis had sought. Assignor estoppel does not bar Davis from challenging the validity of Nancy’s new claims because he could not have made a representation that the broader (and as yet non-existent) claims were valid.
5. Samuel invents a device and his employer, Digital, files an application for his invention. Samuel assigns his invention to Digital. Samuel leaves Digital for a competitor company. After Samuel leaves, Digital files an amendment to the original claims that adds arguably non-patentable subject matter. Assignor estoppel does not bar Samuel from challenging the validity of Digital’s amended claims because he could not have represented the yet-to-be amended claims were valid.

Minerva, like any Supreme Court decision worth its salt, leaves us with some interesting open questions that courts and litigants will need to grapple with in the coming months and years. Here are four open issues that may prove particularly important.

First, to what extent — if at all — can parties alter the default assignor-estoppel doctrine by contract? For example, in hypo 1
above, could Sarah expressly disclaim any representation regarding the validity of the patent and reserve her right to challenge its validity?

Conversely, in hypo 2 above, could Acme simply include a clause in its standard employment agreement whereby Bill agrees not to challenge the validity of any patents that Acme later obtains based on Bill’s inventions? It’s not clear if such clauses would be enforceable.

On one hand, they could permit the kinds of unfair results the majority was concerned about. In modified hypo 1, Sarah would
be permitted to argue that the thing she just sold Wilson was worthless; in modified hypo 2, Bill would be effectively forced to
make a representation about a thing that doesn’t exist yet. On the other hand, is the result really unfair if both parties agree to it in advance?

If parties can simply achieve their desired results via contract notwithstanding assignor estoppel, the implications of this decision may be quite limited. But it’s not clear that this sort of contracting around would be permitted. In other contexts, court have been reluctant to allow parties to avoid background equitable principles via contract.

Consider patent exhaustion, for example: a patentee cannot avoid patent exhaustion by attempting to contractually restrain a
licensee’s ability to dispose of the product. The grant of the license exhausts the patent rights no matter what the contract says.

Second, what does Minerva mean for the Federal Circuit’s rule that assignor estoppel does not apply in inter partes review
proceedings? In Arista Networks Inc. v. Cisco Systems Inc., 908 F.3d 792 (Fed. Cir. 2018) (and in several other cases since), the Federal Circuit held that a patentee cannot raise assignor estoppel in IPR proceedings because the American Invents Act (specifically, 35 U.S.C.A. § 311) allows any “person who is not the owner of a patent” to petition for IPR of a patent.

The Arista court concluded that, even assuming assignor estoppel is a background common-law principle against which Congress is presumed to legislate, this language evidenced congressional intent that assignor estoppel should not apply in IPRs.

Minerva doesn’t necessarily disturb this holding; the court didn’t address Section 311 or assignor estoppel’s application in IPRs. But, given the Minerva majority’s fairly enthusiastic endorsement of assignor estoppel as “well grounded in centuries-old fairness principles,” perhaps the language in Section 311 is not sufficiently explicit to evidence congressional intent to displace the doctrine in the IPR context. We’ll see what the Federal Circuit thinks.

Third, what is the scope of the court’s “change in law” exception to assignor estoppel? Does it require explicit overruling of a prior decision, or is it broader?

This could matter quite a bit, particularly in unsettled areas of the law like patent-eligible subject matter. The Federal Circuit hasn’t actually overruled any of its Section 101 precedents in quite a while, but one could reasonably argue that there have been several Section 101 decisions over the past few years that have “changed the law” as a practical matter.

It is possible that, in assessing the change-in-law exception to assignor estoppel, courts might borrow from the cases addressing the change-in-law exception to forfeiture. But the contours of the change-in-law exception to forfeiture are themselves rather unclear. (For an illustration of this point, check out the cert-stage briefing in Sanofi-Aventis v. Mylan, No. 19-1451.) So importing this body of law might raise more questions than it answers.

Fourth, what does it mean for later claims to be “materially broader” than earlier claims? How much breadth qualifies as “material”?

And, in conducting the “materially broader” inquiry, does one compare the new claims to the old claims, or does one compare the new claims to the disclosure of the assigned application?

The language of the majority opinion in Minerva suggests the former, but that might lead to odd results. To see why, consider an extension of hypo 4 above.

Suppose Davis’s original application describes and enables a quantum computer that can reach processing speeds of 10-100 times existing supercomputers, but the claims in the as-filed application are limited to a 10x processing speed. (Maybe Davis had a bad patent lawyer.)

Nancy buys the application and then seeks a claim directed to a 100x processing speed. Nancy’s new claim is surely “materially broader” than Davis’s original claims, but isn’t it fair to view Davis’ assignment of the application as carrying an implicit representation that a claim to a 100x processing speed would be valid? So why shouldn’t assignor estoppel apply?

On the other hand, if one is supposed to compare the new claims to the disclosure of the assigned application, then the “materially broader” inquiry would effectively collapse into a merits analysis of written description.

In that case, a decision finding assignor estoppel doesn’t apply would also be a decision on the merits: the new claim would be invalid under Section 112. It would be a bit odd to have an equitable doctrine that applies only when the party whom it is supposed to benefit is going to win on the merits anyway.

The Minerva case itself involves a variation of hypo 4, and the Supreme Court has now remanded the case for further analysis by the Federal Circuit, so we may get to see the “materially broader” test in action sooner rather than later. The outcome on remand may provide significant guidance on how the Federal Circuit will apply assignor estoppel in its new and narrower form.

This article also appeared in Volume 28, Issue 7 of Westlaw Journal Intellectual Property.