A parent cannot be invalidated by its own children

How to read the ledger below. Each numbered point carries the kind of evidence behind it. ESTABLISHED is stated in the authority's own text, with a locator. IMPLEMENTED is the rule as the authority actually applied it. OBSERVED is carried by a source outside this one. ASSUMED is the drafter's inference, flagged as such and not the authority speaking. Quotations are checked against a cached copy of the primary source by script.

In one paragraph

The Federal Circuit reversed the District of Delaware in Allergan v. MSN on 13 August 2024 and held that a first-filed, first-issued, later-expiring claim cannot be invalidated for obviousness-type double patenting by a later-filed, later-issued, earlier-expiring reference claim sharing a common priority date, so Allergan kept claim 40 of the first patent ever to cover eluxadoline against two of its own continuations. The court confined In re Cellect to the question of which expiration date to use and said it never decided which claims may serve as a reference. Judge Dyk agreed on that point and dissented from the separate written description holding.

Reader summary

Allergan's '356 patent was the first patent ever to cover eluxadoline. Two later continuations, the '011 and '709 patents, claimed priority from the same application and, having received no patent term adjustment, expired earlier than their own parent, which had received adjustment. The defendants argued that this made the children valid double-patenting references against the parent, and the district court agreed, considering itself bound by In re Cellect.

The Federal Circuit reversed and confined Cellect sharply. Cellect decides which expiration date to use, adjustment included. It does not decide which claims may serve as a reference at all, and the opinion says so twice, adding in a footnote that the patentee in Cellect never raised the reference-propriety question, so party presentation kept the court from reaching it.

On the merits the reasoning is purposive rather than mechanical. The doctrine exists to stop a patentee obtaining a second, later-expiring patent that extends exclusivity on subject matter a first patent already covered. The '356 patent is not a second patent to anything; it is the first. Holding it invalid over its own continuations would invert the doctrine. So the court held that a first-filed, first-issued, later-expiring claim cannot be invalidated by a later-filed, later-issued, earlier-expiring reference claim sharing a common priority date.

Two cautions the headline omits. The panel divided on the separate written-description question, with Judge Dyk dissenting from Part II. And the Office has since read the "first-filed" prong narrowly, against actual filing dates, in a way that keeps this shield away from most pending applications; that is a separate record in this corpus.

Why this matters when you are drafting

This is a shield for the parent, and it does not travel to the child. The protected position is defined by facts fixed years earlier, being first-filed and first-issued in the family. A continuation you are drafting is by construction neither. Reading this case as general protection for a family is the single most common way to misuse it.

Do not let it be read as an exception to the term rules. Cellect still governs which expiration date enters the comparison, adjustment included (claim 8). What this case adds is a limit on which claims may serve as a reference. Those are different questions, and the district court's error was treating the first as answering the second.

The parent's adjustment is no longer automatically a liability, in this one configuration. After Cellect alone, adjustment on a parent looked like pure exposure whenever a continuation expired earlier. This case removes that specific exposure where the parent is first-filed and first-issued. It does not remove it elsewhere, and the interaction with the Office's own reading is where practitioners get hurt.

Check the Office's position before relying on this in prosecution. The USPTO's Appeals Review Panel has since read the "first-filed" prong as keyed to actual filing dates and "common priority date" as keyed to patent term filing dates, and told examiners the exception will rarely if ever arise during original examination. That is the ex-parte-baurin record in this corpus, and it is the difference between how this case reads in litigation and how it operates at the Office.

Note what the case did not have to decide. Claim 11: filing order and issuance order agreed on these facts. A family where the first-filed member is not the first-issued is unresolved by this opinion.

The dissent is about continuation drafting, and it is worth reading for that alone. Claim 14 describes continuations filed after litigation began, claiming formulations without a limitation the allowed claims had required. Whatever one makes of the outcome, that fact pattern is a live drafting decision, and one judge thought it failed.

What the authority establishes

[ESTABLISHED, the question and the answer, stated together at the outset] "can a first-filed, first-issued, later-expiring claim be invalidated by a later-filed, later-issued, earlier-expiring reference claim having a common priority date? We hold that it cannot." (P. 11.)

[ESTABLISHED, the holding restated at the end of the analysis] "We therefore hold that a first-filed, first-issued, later-expiring claim cannot be invalidated by a later-filed, later-issued, earlier-expiring reference claim having a common priority date." (P. 17.)

[ESTABLISHED, the principle underneath it] "As the first-filed, first-issued patent in its family, it is the patent that sets the maximum period of exclusivity for the claimed subject matter and any patentably indistinct variants." (P. 17.)

[ESTABLISHED, the doctrinal purpose that drives the result] "The doctrine's primary goal is to prevent an unjustified timewise extension of patent exclusivity beyond the life of a patent." (P. 11.) The court reasons the '356 patent cannot extend exclusivity on subject matter it was itself the first to cover.

[ESTABLISHED, the disposition] "we reverse the district court's judgment of invalidity of the asserted claims under 35 U.S.C. § 112 and obviousness-type double patenting". (P. 30.) Both grounds fell.

[ESTABLISHED, and this is what practitioners most often get wrong about the case] Cellect is confined, not extended: "Cellect answered a different question than that at issue here" (P. 15.), and "Cellect does not address, let alone resolve, any variation of the question presented here". (P. 15.)

[ESTABLISHED, why Cellect could not have decided it] In footnote 6 the court adds that "Cellect did not involve the situation presented here of ODP with respect to a first-filed, first-issued patent" (P. 15.), and that the patentee there never challenged whether the reference claims were proper references, so "under the principle of party presentation, the court did not consider that issue". (P. 15.)

[IMPLEMENTED] Cellect still binds on the one thing it decided. The court applies the '356 patent's adjusted 2026 expiration date rather than the 2025 date it would have shared with the references, and then holds that expiring later "is of no consequence here". Which date to use and whether a claim is a proper reference are separate questions, and only the first was settled. (Pp. 15 to 17.)

[ESTABLISHED, the specific conclusion on these facts] "the claims of the '011 and '709 reference patents are not proper ODP references that can be used to invalidate claim 40 of the '356 patent". (P. 16.)

[ESTABLISHED, the structural background that makes the problem possible] After the 1994 term change, "post-URAA, there is little risk of an unjustified extension of term subject to ODP because all patents to an invention that share a priority date are expected to expire on the same day", but "a common priority date does not always guarantee a common expiration date". (Pp. 12 to 13.) Adjustment is what breaks the alignment.

[ESTABLISHED, both measures point the same way here] The court stresses the '356 patent is first "whether we measure by filing date or by issuance date", and that neither continuation application "was even filed until after the '356 patent issued". (P. 16.) On these facts the two measures agree, so the opinion had no occasion to say which controls when they diverge.

[ESTABLISHED, written description, read at framing level only] Part II reverses the written-description invalidation, and the court sets it apart from the usual case at the outset: "this is not a typical written description case", and not a blaze-marks case where claims recite a species and the specification describes only a genus. (P. 21.)

[ESTABLISHED, the panel divided] Judge Dyk concurred on double patenting but wrote: "I respectfully dissent from Part II of the majority's opinion concerning written description". (P. 31.) The double-patenting holding in claims 1 through 11 is therefore unanimous; the written-description holding is not.

[ESTABLISHED, the dissent's characterization of the drafting sequence] "After litigation commenced, Allergan filed new continuation applications claiming formulations that do not require a glidant". (P. 32.) That sequence, continuations drafted around a limitation once litigation was under way, is the fact the dissent treats as central.

[ASSUMED, drafter's inference, flagged as such] Claims 1 through 3 protect a position a practitioner occupies rather than one they can draft into. Nothing in a continuation being drafted today can make it first-filed or first-issued in its own family. The opinion does not frame it this way; it follows from the holding's own terms.

Quotations, verified against the source

From the majority unless marked. Page locators computed from the docket stamp, which runs continuously through the dissent.

"The doctrine's primary goal is to prevent an unjustified timewise extension of patent exclusivity beyond the life of a patent" (P. 11.)
"can a first-filed, first-issued, later-expiring claim be invalidated by a later-filed, later-issued, earlier-expiring reference claim having a common priority date" (P. 11.)
"a common priority date does not always guarantee a common expiration date" (P. 13.)
"Cellect answered a different question than that at issue here" (P. 15.)
"Cellect does not address, let alone resolve, any variation of the question presented here" (P. 15.)
"under the principle of party presentation, the court did not consider that issue" (P. 15.)
"Cellect did not involve the situation presented here of ODP with respect to a first-filed, first-issued patent" (P. 15.)
"the claims of the '011 and '709 reference patents are not proper ODP references that can be used to invalidate claim 40 of the '356 patent" (P. 16.)
"We therefore hold that a first-filed, first-issued, later-expiring claim cannot be invalidated by a later-filed, later-issued, earlier-expiring reference claim having a common priority date" (P. 17.)
"As the first-filed, first-issued patent in its family, it is the patent that sets the maximum period of exclusivity for the claimed subject matter and any patentably indistinct variants" (P. 17.)
"this is not a typical written description case" (P. 21.)
"we reverse the district court's judgment of invalidity of the asserted claims under 35 U.S.C. § 112 and obviousness-type double patenting" (P. 30.)
Dyk, J., concurring in part and dissenting in part: "I respectfully dissent from Part II of the majority's opinion concerning written description" (P. 31.)
Dyk, J.: "After litigation commenced, Allergan filed new continuation applications claiming formulations that do not require a glidant" (P. 32.)

Tensions and open questions

Filing order versus issuance order. The holding names both prongs, and on these facts they coincided (claim 11). A family whose first-filed member is not its first-issued member is not addressed.

What "common priority date" means is left to later readers. The phrase carries real weight in the holding, and this opinion does not define whether it means the shared priority claim, the patent term filing date, or something else. The Office later read it as patent term filing date, which is a reading, not this court's word.

How much of Cellect survives is stated negatively. The court says what Cellect did not decide (claims 6 and 7) more clearly than it says what Cellect still requires. Claim 8 is this record's reading of the residue, tiered IMPLEMENTED for that reason.

The written-description half is genuinely contested. A unanimous holding and a divided one sit in the same opinion. Citing Allergan v. MSN without saying which part is imprecise in a way that matters.

What this record does and does not cover

Discussion Part I, the double-patenting holding, was read end to end including footnote 6, which is the load-bearing footnote and the part most likely to be skipped.

What was NOT read closely, and is therefore outside what this record can support: the substance of Part II's written-description analysis and the dissent's counter-analysis. This record states that Part II reversed, that it distinguishes itself from a blaze-marks case, that Judge Dyk dissented from it, and how the dissent characterizes the drafting sequence. It cannot support any proposition about why the majority found the description adequate, or about the standard applied. A reader wanting that must read pages 21 to 35 directly. The district court opinion, the patents themselves, and the trial record are known here only through this court's characterization.

Currency

This record was verified against its primary source on 2026-08-15. That is a fidelity check: the quotations, locators and attribution are accurate to that source. Whether the authority has since been narrowed, distinguished, abrogated or reversed has not been checked here. Confirm it still stands before relying on it.