Allergan protects less than practitioners hoped

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 USPTO Appeals Review Panel reversed its own Board in Ex parte Baurin on 6 August 2026 and reinstated obviousness-type double patenting rejections, holding that Allergan shields only a claim that is first-filed by actual filing date, first-issued, and later-expiring inside one family sharing a patent term filing date, and that the risk of split ownership alone can sustain a rejection even where no patent term is extended. USPTO personnel are told to keep applying pre-Allergan MPEP 804 practice because the Allergan exception will rarely if ever arise in original examination.

Reader summary

Sanofi's application 17/135,529 claims priority through intervening continuations to a 2012 non-provisional, so any patent issuing from it expires 2032-03-28. The reference the Examiner used against it, the '922 patent, was actually filed in 2017 and expires 2037, five years later. The application, in other words, could not possibly extend the reference's exclusivity: it dies first.

The Board found that decisive and reversed the Examiner. The Appeals Review Panel, convened by the Director, reversed the Board and reinstated the rejections.

Two moves did the work. First, Allergan v. MSN does not apply, because each of its three prongs fails on these facts, and the panel read each prong precisely: "first-filed" means actual filing dates, "first-issued" cannot describe a still-pending application, and "common priority date" means patent term filing dates, which these two do not share. Second, and more consequentially, the panel held that improper term extension is not the only rationale for obviousness-type double patenting. The anti-harassment rationale, preventing multiple assignees from suing separately on obvious variants, can sustain a rejection on its own, even where no term is extended at all.

The decision is unusual in that its own authors disagree with the rule they applied. Part III.C says that were the Office not bound by Federal Circuit precedent, this panel would clarify that hypothetical harassment "generally should not form a standalone basis" for rejections during examination, and it sketches a replacement framework. That framework is a proposal contingent on the Federal Circuit saying so, not law, and the decision says the Office would welcome exactly that clarification. The question is already before the court.

Why this matters when you are drafting

The shield you may be counting on is narrower than its headline. After Allergan it became common to describe the parent as safe from its own children. This decision says the protection attaches to a claim that is first-filed by actual filing date, first-issued, and later-expiring, in a family sharing a patent term filing date. A continuation being drafted today is, by construction, not first-filed and not first-issued. The shield does not travel to it.

The branch-survival question changed shape. The intuition that a branch is safe when it expires before its reference is exactly the Board's reasoning here, and it was reversed. On this decision an earlier-expiring pending claim can still be rejected over a later-filed, later-expiring reference, because the rejection can rest on split-ownership risk alone. When you are deciding whether a branch survives, term arithmetic is no longer sufficient.

Watch the two different filing dates. The decision turns on a distinction that is easy to collapse: the actual filing date, and the patent term filing date that starts the twenty-year clock and generally excludes foreign and provisional priority. One Allergan prong keys on each. The panel says the Board's own error was comparing the wrong one (claim 12), which is a useful warning that this is a mistake a tribunal actually made.

Terminal disclaimers carry a cost the term math does not show. The common-ownership provision (claim 11) means a disclaimer filed to clear a rejection also constrains who may own and enforce the patent later. The panel's departed-inventor hypothetical (claim 18) is worth reading before advising a client with collaborators, spin-outs, or assigned improvement filings.

Treat the framework in Part III.C as a forecast, not a rule. It is explicitly contingent on Federal Circuit clarification. Advising on it as current practice would be an error of posture. The one thing it does tell you is where the Office would go if the court narrows the anti-harassment rationale, and In re Ablynx (claim 19) is the docket to watch.

What the authority establishes

1. [ESTABLISHED, the disposition] The panel reversed the Board and reinstated the Examiner's rejections: "we reverse the Board's Appeal Decision and Rehearing Decision. The Examiner's rejections of record for obviousness-type double patenting, therefore, are affirmed." (P. 31.) Announced earlier as "On review, we REVERSE." (P. 2.)

1. [ESTABLISHED, the facts that make this case unlike Allergan] The application claims priority through intervening continuations to a non-provisional filed 2012-03-28, so its term ends 2032-03-28. The reference '922 patent has an actual and patent term filing date of 2017-04-13, issued 2021-01-05, and runs to 2037-04-13 plus 70 days of patent term adjustment. The challenged claims therefore expire roughly five years BEFORE the reference. (Pp. 3 to 4.)

1. [ESTABLISHED] Obviousness over the reference was never contested. The dispute was only whether the reference could serve as a double-patenting reference at all. (P. 4.) The panel notes the asymmetry: the pending claims would have been obvious over the reference's claims, but the reference's claims would not have been obvious over the application's published disclosure. (P. 4.)

1. [ESTABLISHED, the Allergan holding as restated] Allergan's rule is "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. 12.)

1. [ESTABLISHED, and this is the prong practitioners most often assume] "The 'first-filed' prong refers to actual filing dates" (P. 12.), not patent term filing dates and not priority dates. Here that prong fails because "the '529 Application does not have the first actual filing in its family". (P. 13.)

1. [ESTABLISHED] The "first-issued" prong cannot be satisfied by a pending application: "the claims of the '529 Application are not 'first-issued' because the '529 Application is still pending". (P. 13.)

1. [ESTABLISHED, and note it points the OTHER way from claim 5] "The 'common priority date' prong refers to patent term filing dates" (P. 13.), and it fails here because the application and the reference have different ones. So a single holding uses actual filing dates for one prong and patent term filing dates for another; conflating them is how the Board went wrong, per claim 12.

1. [IMPLEMENTED] All three prongs are treated as independently necessary. The panel walks each one and finds each unsatisfied before concluding "Thus, Allergan's holding does not apply here." (P. 13.) Nothing suggests a prong can be excused.

1. [ESTABLISHED, the pivotal holding] "Improper term extension, however, is not the only rationale underpinning the OTDP doctrine." (P. 14.) The panel concludes "that the Board erred in dismissing the anti-harassment rationale as inadequate to support the Examiner's OTDP rejections". (P. 17.)

1. [ESTABLISHED] The anti-harassment rationale is grounded on Fallaux, Hubbell, Van Ornum and Cellect, and the panel stresses that in Fallaux and Hubbell there was no term-extension concern at all, so reading those discussions as dicta "would mean that the rejections were affirmed for no underlying OTDP rationale at all." (Pp. 19 to 20.)

1. [ESTABLISHED] The regulatory hook is the common-ownership provision: a terminal disclaimer must make the patent "enforceable only for and during such period that said patent is commonly owned with the application or patent which formed the basis for the judicially created double patenting". (Pp. 15 to 16.) That is what makes the anti-harassment rationale operative rather than rhetorical.

1. [ESTABLISHED, a correction the panel makes to the Board] The Board thought term extension was at issue in Hubbell; the panel says "The Board is mistaken," and locates the error in "its reliance on and comparison of actual filing dates, not patent term filing dates for determining the terms." (P. 19 n.7.)

1. [ESTABLISHED, the operative instruction to examiners, Part V] "USPTO personnel should continue to follow existing pre-Allergan OTDP practice discussed in MPEP § 804" unless both (1) the application and the reference are in the same family with the same patent term filing date, and (2) the claims under examination are first-filed, first-issued and later-expiring. (P. 31.)

1. [ESTABLISHED, and this is the sentence that governs day-to-day prosecution] The Allergan exception "will rarely, if ever, arise in original examination because a patent examiner will rarely, if ever, be in a position to determine that the claims under examination will be first-filed, first-issued, and later-expiring claims." (P. 32.) The panel says it can arise in reexamination or reissue. (P. 32.)

1. [ESTABLISHED, patent term extension is still untouched] Double patenting may rest on expiration differences caused by patent term adjustment, but "it cannot be based on different expiration dates arising from PTE under § 156", citing Novartis v. Ezra. (P. 23 n.8.)

1. [ESTABLISHED, PROPOSAL ONLY, NOT THE HOLDING] Part III.C sketches a framework the Office would adopt only "should the Federal Circuit clarify that OTDP rejections cannot be based solely on the anti-harassment rationale." (P. 21.) Across families it would ask one question: "Does a putative OTDP reference have a later patent term filing date than the patent term filing date of the application under examination?" and if yes, the term-extension analysis "ends." (P. 28.) Within a family it would use actual filing dates, so later-filed progeny can be rejected over earlier-filed parents but "such rejections would not work in reverse." (P. 30.)

1. [ESTABLISHED, the panel's own disagreement with the rule it applied] "Were the Office not bound by the Federal Circuit's precedents applying the OTDP anti-harassment rationale, this ARP would clarify that the risk of separate ownership and hypothetical harassment generally should not form a standalone basis for OTDP rejections during examination." (P. 24.) It calls the concern "more theory than real," quoting Judge Rich that "we do not see the courts bogged down with harassment suits". (P. 25.)

1. [ESTABLISHED, the practical harm the panel identifies] A standalone anti-harassment rationale "potentially sets a 'trap' with no viable escape for at least some applicants" (P. 27.), illustrated by a departed-inventor hypothetical in which a foundational application is rejected over a former colleague's later improvement patent and cannot cure the rejection without acquiring that patent. (Pp. 26 to 27.)

1. [ESTABLISHED, an open appellate question with a docket] The panel says "the Office would welcome that clarification from the court" and states the issue "is now before the court in In re: Ablynx N.V., Appeal No. 26-1333 (sub nom Ex parte Baumeister)", citing Ex parte Baumeister, 2025 WL 3515282 (PTAB Nov. 20, 2025). (P. 20.)

1. [ESTABLISHED, examination should not speculate] Under the proposed framework the Office would rely on "known facts of record, not speculation on unknown matters such as hypothetical future grants of PTA, possible withdrawal of terminal disclaimers, or potential priority benefit changes." (P. 28 n.10.)

1. [ASSUMED, drafter's inference, flagged as such] Claims 13 and 14 together read as an instruction that examiners should almost never apply Allergan affirmatively, which means a practitioner invoking Allergan in prosecution is asking an examiner to do the thing this decision tells them they will rarely be positioned to do. The decision does not say this in terms; it is the drafter's reading of Part V's effect.

Quotations, verified against the source

All from the Appeals Review Panel's decision. Page locators computed from the printed pagination of the cached document.

• "On review, we REVERSE." (P. 2.)
• "The 'first-filed' prong refers to actual filing dates" (P. 12.)
• "Thus, Allergan's holding does not apply here" (P. 13.)
• "Improper term extension, however, is not the only rationale underpinning the OTDP doctrine" (P. 14.)
• "We conclude that the Board erred in dismissing the anti-harassment rationale as inadequate" (P. 17.)
• "the Office would welcome that clarification from the court" (P. 20.)
• "Does a putative OTDP reference have a later patent term filing date than the patent term filing date of the application under examination" (P. 28.)
• "such rejections would not work in reverse" (P. 30.)
• "USPTO personnel should continue to follow existing pre-Allergan OTDP practice discussed in MPEP § 804" (P. 31.)
• "It will rarely, if ever, arise in original examination" (P. 32.)

Tensions and open questions

The decision argues against itself, deliberately. Part III.B applies the anti-harassment rationale because precedent compels it; Part III.C says the panel would abandon it if it could. Both are the same authors on the same day. A reader who quotes only one half misrepresents the decision, and the halves point opposite ways for planning purposes.

Is anti-harassment truly freestanding, or is this the Office's reading of precedent? The panel says Fallaux and Hubbell affirmed rejections where no term extension existed, so the rationale must be doing independent work. The Board read those passages as dicta. The Federal Circuit has not resolved it, which is why claim 19 matters.

How binding is this outside the Office? It governs USPTO personnel. It is not court precedent, and a district court or the Federal Circuit could reject its reading of Allergan wholesale. Its practical force during prosecution is nonetheless immediate.

Unresolved by this record: what "same family" means for Part V's first condition. The decision uses the term without defining its boundary for divisional and continuation-in-part branches. Nothing in the text read here settles it.

What this record does and does not cover

The decision was read end to end, all 34 pages, including Part III.C and Part V, which are the parts most likely to be skipped and are where the practitioner-relevant guidance sits. Footnotes 1 through 11 were read; footnotes 3, 7, 8 and 10 are in the ledger.

What was NOT read closely, and is therefore outside what this record can support: the substance of the eleven amicus briefs beyond the two passages the decision quotes, and the underlying Board Appeal Decision and Rehearing Decision as separate documents. This record describes what the panel said about them, not what they say. Claims about the dissent in the Board's Rehearing Decision are outside this record entirely.

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.