Disclosing every possibility describes none of them

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 affirmed the Patent Trial and Appeal Board in Regents of the University of Minnesota v. Gilead Sciences on 6 March 2023 and held that earlier applications reciting a compendium of possible chemical substituents for every position did not describe the narrower subgenus claimed later, because a laundry list of every possibility would otherwise describe every species inside it. The patent therefore lost its claimed priority date and the challenged claims were anticipated by a reference published in the interval.

Reader summary

Minnesota's patent claimed a subgenus of nucleoside compounds. Its priority chain ran back through several earlier applications, and the question was whether any of them described the subgenus well enough that the later claims could claim their filing date. If not, an intervening publication became prior art and the claims were anticipated.

The earlier applications disclosed a great deal. They recited common organic functional groups for each substituent position, so that the claimed subgenus was formally within what the combinations could produce. Minnesota argued that this was literal description, or at least provided blaze marks to the subgenus in its broad outlines.

The court disagreed, and its reasoning is the useful part. If reciting every possible moiety for every position counted as describing every species those recitations could generate, then a laundry list would describe everything inside it. Written description asks whether the earlier document conveys possession of the particular thing later claimed, and a combinatorial space does not convey possession of each combination within it.

The consequence was not a narrow one. Losing the priority date exposed the claims to a reference published in the interval, and they were anticipated. This is the shape practitioners should carry from Tier B: a written-description defect in a priority document is rarely survivable on its own terms, because what it costs is the date, and the date is what was keeping the art out.

Why this matters when you are drafting

This is the central failure mode for a continuation claiming a selection. The parent discloses a space; the continuation claims a point or a region inside it. Claim 4 is why that does not work by containment alone. What the later claim needs is something in the earlier document that points at the selection.

Write blaze marks while you still can. The only moment you can add direction to a priority document is before it is filed. Preferred embodiments, worked examples, stated combinations and explicit narrowing all function as marks. Claim 3 is the test they have to meet: do they guide attention to what was later claimed.

Enumeration is not insurance, and may be the opposite. Claim 10 is the honest inference, and it cuts against a common drafting instinct. Listing every substituent to keep options open produces exactly the laundry list the court treats as describing nothing in particular.

Losing priority is rarely the whole injury. Claim 1: the date fell and anticipation followed immediately. When assessing a support risk in a priority document, ask what published in the interval, because that is the real exposure.

Know which side of the line your art sits on. rai-strategic-v-philip-morris reached the opposite result for a numeric subrange in a predictable, mechanically simple art. This case is a combinatorial chemical space. The written-description standard is the same; the amount of direction it demands is not.

What the authority establishes

1. [ESTABLISHED, the disposition and what it cost] "we affirm the Board's final written decision holding that NP2 and P1 do not provide sufficient written description to support the '830 patent claims, and that Sofia therefore anticipated these claims". (P. 14.) Support failed, priority fell, anticipation followed.

1. [ESTABLISHED, the two routes that were tried and failed] The Board found the earlier documents "contained neither ipsis verbis support nor sufficient blaze marks to guide the skilled artisan to the claims of the '830 patent". (P. 5.) Literal description and blaze marks are alternatives, and a claim needs one of them.

1. [ESTABLISHED, what blaze marks are for] "For genus claims, which are present here, we have looked for blaze marks within the disclosure that guide attention to the claimed species or subgenus." (P. 7.) The metaphor is directional: the disclosure must point, not merely contain.

1. [ESTABLISHED, the reductio that decides the case] "Were this the case, a 'laundry list' disclosure of every possible moiety for every possible position would constitute a written description of every species in the genus." (P. 9.) That consequence is treated as unacceptable, which is what defeats the containment argument.

1. [ESTABLISHED, what the earlier applications actually disclosed] "The claims of P1 recite a compendium of common organic chemical functional groups, yielding a laundry list disclosure of different moieties for every possible side chain or functional group." (P. 9.) Breadth of disclosure was never in doubt; direction was.

1. [ESTABLISHED, the argument the patentee made] Minnesota asserted its earlier applications "literally described, or provided blaze marks to, the subgenus of the '830 claims in its broad outlines". (P. 6.) "Broad outlines" is the theory the court rejects.

1. [ESTABLISHED, the misreading of Ariad the court corrects] The court says Minnesota "mischaracterizes Ariad" by reading it to mean that disclosing structural features common to the members of a genus "demonstrates possession of, and thereby supports, the claimed genus". (P. 10.) Common structural features are not by themselves possession. (The opinion nests a quotation inside a quotation here, so this claim quotes the two unnested spans and paraphrases the inner one.)

1. [ESTABLISHED, the standard of review, which matters to how far this travels] "We thus review the Board's decision regarding written description for substantial evidence." (P. 7.) A finding is supported "if a reasonable mind might accept the evidence as adequate to support the finding". (P. 5.)

1. [IMPLEMENTED, the priority chain was evaluated document by document] The Board looked at each application separately, finding one was filed after the reference published and another carried the same disclosure as its predecessor. (P. 4.) A chain is only as good as the specific document that must carry the description.

1. [ASSUMED, drafter's inference, flagged as such] Claims 4 and 5 together mean breadth of disclosure can be actively unhelpful: the more exhaustively a specification enumerates options, the more it resembles a laundry list and the weaker the argument that any particular combination was possessed. The opinion does not say enumeration hurts; it follows from the reductio being the reason containment fails.

Quotations, verified against the source

All from the Federal Circuit's opinion. Page locators computed from the docket stamp.

• "A finding is supported by substantial evidence if a reasonable mind might accept the evidence as adequate to support the finding" (P. 5.)
• "contained neither ipsis verbis support nor sufficient blaze marks to guide the skilled artisan to the claims of the '830 patent" (P. 5.)
• "literally described, or provided blaze marks to, the subgenus of the '830 claims in its broad outlines" (P. 6.)
• "For genus claims, which are present here, we have looked for blaze marks within the disclosure that guide attention to the claimed species or subgenus" (P. 7.)
• "We thus review the Board's decision regarding written description for substantial evidence" (P. 7.)
• "Were this the case, a 'laundry list' disclosure of every possible moiety for every possible position would constitute a written description of every species in the genus" (P. 9.)
• "The claims of P1 recite a compendium of common organic chemical functional groups, yielding a laundry list disclosure of different moieties for every possible side chain or functional group" (P. 9.)
• "mischaracterizes Ariad" (P. 10.)
• "demonstrates possession of, and thereby supports, the claimed genus" (P. 10.)
• "we affirm the Board's final written decision holding that NP2 and P1 do not provide sufficient written description to support the '830 patent claims, and that Sofia therefore anticipated these claims" (P. 14.)

Tensions and open questions

How many blaze marks are enough is not stated, and cannot be. Claim 3 gives the function and claim 8 explains why the answer stays fact-bound: this is substantial-evidence review of a Board finding, so the opinion affirms a conclusion rather than setting a threshold.

The line between a laundry list and a genuine genus disclosure is drawn by example. Claim 5 describes what failed. Nothing here says what a sufficient chemical genus disclosure looks like, which makes the positive counterexample in this tier a numeric range rather than a chemical genus.

Ariad is the disputed authority. Claim 7 corrects a reading rather than announcing a new rule, so the holding's reach depends on how Ariad is understood generally, and this opinion addresses only what it does not hold.

Unresolved by this record: whether a later-added blaze mark can help. Everything here turns on what the earlier documents contained. Nothing addresses continuation-in-part practice, where new matter is added and priority splits by claim.

What this record does and does not cover

The written-description and priority analysis was read end to end, because it is the reason this record exists.

What was NOT read closely, and is therefore outside what this record can support: the chemical substance of the claimed and disclosed compounds, the anticipation analysis, and the expert testimony. This record says nothing about whether the Sofia reference discloses the claims, only that anticipation followed once priority was lost. The Board decision and the priority applications are known here only through the 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.