The specification did not change. The claims did.

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 district court in Biogen v. Mylan on 30 November 2021 and held claims to a specific therapeutically effective dose invalid for lack of written description. Biogen left its 2007 specification unchanged and in 2011 wrote entirely new claims to a dose the Phase III results had just validated, which preserved the 2007 priority date but meant the claims had to find support in a drug discovery disclosure that mentioned the dose once, at the end of one range among several. Later proof of efficacy does not count, and review was for clear error, so the decision turns on this trial record rather than stating a rule about single mentions. Judge O Malley dissented.

Reader summary

Biogen filed a provisional in February 2007 whose disclosure was about screening compounds for activation of a biological pathway. In 2011, after Phase III results showed a 480 mg daily dose of dimethyl fumarate was effective for multiple sclerosis, it amended the title, added an inventor, wrote entirely new claims to that dose, and left the specification alone. That combination kept the 2007 priority date on the face of the file. It also framed the question the court decided.

The disclosure did mention the number. Once. At the end of one range among several, alongside ranges that also encompassed doses everyone agreed were ineffective and doses well above the one known to work. The district court found that a skilled artisan reading the 2007 document would not have recognized the 480 mg dose as therapeutically effective for multiple sclerosis, and the Federal Circuit found no clear error.

Two features of the reasoning matter more than the chemistry. First, later proof of efficacy is irrelevant: the question is what the specification conveyed at the time of filing. Second, the inventor's own testimony that his work was exploratory and was never aimed at clinical dosing was treated as evidence that possession was absent.

Biogen's rescue argument was that the 480 mg dose was anchored to the known-effective 720 mg dose by the disclosed range. The court answered that the same sentence anchors an ineffective dose just as firmly. A range that reaches your limitation also reaches the limitations that would sink you.

Judge O'Malley dissented, and the dissent's footnote is the honest counterweight: the dose was expressly disclosed, and no case law requires a claim element to be recited more than once.

Why this matters when you are drafting

An unchanged specification is the point of the strategy and the source of the exposure. Claim 3. The reason to leave it alone is to keep the priority date. The consequence is that the new claim must find its support in a document written before anyone knew what to claim.

A single mention inside a range is the weakest form of support there is. Claims 6 and 8. A laundry list of ranges reaches your limitation and every neighbouring one, and the court will say so. When drafting a parent, a value you may later want to claim should appear independently, not only as an endpoint. When drafting a continuation, check whether the value you are about to claim ever appears outside a range.

What you learned after filing cannot be read back into the specification. Claim 9. Clinical results, later testing, market data: none of it moves the possession inquiry. The temptation in continuation practice is exactly the reverse, because the later knowledge is what motivated the branch.

Claiming an effect imports a disclosure burden for that effect. Claim 10. A method claim reciting a therapeutically effective amount is a claim to a result, and the specification must support the result, not merely the compound.

Inventor testimony is where possession is lost. Claim 13. The named inventor of the parent described his own work as exploratory and not directed at the claimed use, which is a predictable answer when a continuation claims something outside the parent's original purpose.

Read this record with its dissent. The Tensions section carries it. The dose was expressly disclosed, and the dissent says no authority requires repetition. That limit belongs with the holding whenever this case is cited.

What the authority establishes

[ESTABLISHED, the disposition] "we affirm the district court's decision that Mylan satisfied its burden of showing, by clear and convincing evidence, that the asserted '514 Patent claims are invalid for lack of written" description under section 112. (P. 21.) The sentence is split at the page break; the remainder is on the next page and this record quotes only the part that verifies as one span.

[ESTABLISHED, the finding in one sentence] "the record shows that the inventors were not in possession of a method of administering a therapeutically effective dose of DMF480 to treat MS on or before the February 8, 2007 priority date." (P. 22.)

[ESTABLISHED, and this is the mechanic a continuation practitioner should recognize] Besides amendments to inventorship and title, Biogen made no other changes to the specification. "This enabled Biogen to claim a priority date of February 8, 2007, despite filing wholly new claims alongside the amendments." (P. 10.) The court states it neutrally. It is what preserved the date and what framed the defect.

[ESTABLISHED, the question in the court's own framing] "The core issue in this appeal is whether the specification Biogen filed on February 8, 2007 supports the 2011 claims that issued in the '514 Patent." (P. 14.) Old specification, new claims, four years apart.

[ESTABLISHED, where possession must be shown] The district court properly noted that "it is the specification itself that must demonstrate possession." (P. 15.) Possession is an "objective inquiry into the four corners of the specification from the perspective of a" skilled artisan. (P. 14.)

[ESTABLISHED, the fact the case turned on] "The DMF480 dose is listed only once in the entire specification." (P. 16.) And its single appearance was not standalone: it "appears at the end of one range among a series of ranges". (P. 16.)

[ESTABLISHED, the conclusion drawn from that] "the specification's focus on basic research and broad DMF-dosage ranges show that the inventors did not possess a therapeutically effective DMF480 dose at the time of filing in 2007." (P. 16.)

[ESTABLISHED, why the range argument failed, and it is the transferable point] Biogen argued the claimed dose was anchored to the known-effective one by the disclosed range. "Not only does the specification anchor an ineffective dose, it also expands the purported range of therapeutic efficacy" beyond the effective dose in both directions. (P. 18.) A range that supports your limitation supports the ones that defeat it equally.

[ESTABLISHED, later evidence does not help] "That Biogen later established the therapeutic efficacy of DMF480 is of no import to the written-description analysis." (P. 17.) The inquiry is fixed at the filing date.

[ESTABLISHED, the rule about claimed results] An inventor need not "prove that a claimed pharmaceutical compound actually achieves a certain result. But when the inventor expressly claims that result, our case law provides that" the result must be supported by adequate disclosure in the specification. (P. 17.) Claiming an effect imports a disclosure burden for that effect.

[ESTABLISHED, research disclosures do not carry treatment claims] "the law is clear that a patent cannot be awarded for mere theoretical research without more" (P. 18.), because "The written-description requirement limits patent protection only to individuals who perform the difficult work of producing a complete and final invention featuring all its claimed limitations and publicly disclose the fruits of that effort." (P. 18.)

[ESTABLISHED, the actual finding, stated as a finding] The district court did not clearly err in finding that "a skilled artisan would not have recognized, based on the single passing reference to a DMF480 dose in the disclosure, that DMF480 would have been efficacious in the treatment of MS". (P. 17.)

[ESTABLISHED, inventor testimony again supplies the evidence] The original named inventor "denied that his research could be extrapolated to a clinical dose of DMF" (P. 10.), and Biogen's expert opinion that a skilled artisan would be drawn to the dose was found "neither credible nor persuasive." (P. 19.)

[IMPLEMENTED, and it governs how far this case travels] This is clear-error review of a bench-trial finding under a clear-and-convincing burden, and the panel expressly declined to disturb credibility assessments. Claims 1 and 13. The holding is that this record supported this finding, not that a once-mentioned limitation is unsupported as a matter of law.

[ESTABLISHED, the fact-dependence the court states directly] "the written-description analysis is highly dependent on the facts of each case" (P. 14.), and "written description is judged based on the state of the art as of the priority date" (P. 14.).

[ASSUMED, drafter's inference, flagged as such] The shape here is the ordinary late continuation: hold a chain open, learn what matters commercially, then claim it from the original disclosure. Claims 3, 6 and 9 are what that shape costs when the original disclosure was written for a different purpose. The opinion does not describe it as a continuation-strategy case; this follows from those claims read together with the prosecution history in claim 3.

Quotations, verified against the source

All from the Federal Circuit's opinion. Page locators computed from the docket stamp, which numbers the majority and the dissent in one continuous sequence.

"This enabled Biogen to claim a priority date of February 8, 2007, despite filing wholly new claims alongside the amendments." (P. 10.)
"denied that his research could be extrapolated to a clinical dose of DMF" (P. 10.)
"The core issue in this appeal is whether the specification Biogen filed on February 8, 2007 supports the 2011 claims that issued in the '514 Patent." (P. 14.)
"objective inquiry into the four corners of the specification from the perspective of a" (P. 14.)
"the written-description analysis is highly dependent on the facts of each case" (P. 14.)
"written description is judged based on the state of the art as of the priority date" (P. 14.)
"it is the specification itself that must demonstrate possession." (P. 15.)
"The DMF480 dose is listed only once in the entire specification." (P. 16.)
"appears at the end of one range among a series of ranges" (P. 16.)
"the specification's focus on basic research and broad DMF-dosage ranges show that the inventors did not possess a therapeutically effective DMF480 dose at the time of filing in 2007." (P. 16.)
"That Biogen later established the therapeutic efficacy of DMF480 is of no import to the written-description analysis." (P. 17.)
"prove that a claimed pharmaceutical compound actually achieves a certain result. But when the inventor expressly claims that result, our case law provides that" (P. 17.)
"a skilled artisan would not have recognized, based on the single passing reference to a DMF480 dose in the disclosure, that DMF480 would have been efficacious in the treatment of MS" (P. 17.)
"the law is clear that a patent cannot be awarded for mere theoretical research without more" (P. 18.)
"The written-description requirement limits patent protection only to individuals who perform the difficult work of producing a complete and final invention featuring all its claimed limitations and publicly disclose the fruits of that effort." (P. 18.)
"Not only does the specification anchor an ineffective dose, it also expands the purported range of therapeutic efficacy" (P. 18.)
"neither credible nor persuasive." (P. 19.)
"we affirm the district court's decision that Mylan satisfied its burden of showing, by clear and convincing evidence, that the asserted '514 Patent claims are invalid for lack of written" (P. 21.)
"the record shows that the inventors were not in possession of a method of administering a therapeutically effective dose of DMF480 to treat MS on or before the February 8, 2007 priority date." (P. 22.)
"The majority does not, and cannot, deny that the claimed DMF480 dose is expressly disclosed." (P. 32.)
"the majority cites no case law (and I know of none) for the proposition that the written description requirement demands that a patentee recite a claim element repeatedly to pass written description muster." (P. 32.)

Tensions and open questions

The dissent's footnote is the strongest limit on this case and it comes from inside the opinion. Judge O'Malley wrote that "The majority does not, and cannot, deny that the claimed DMF480 dose is expressly disclosed." (P. 32.) and that "the majority cites no case law (and I know of none) for the proposition that the written description requirement demands that a patentee recite a claim element repeatedly to pass written description muster." (P. 32.) Anyone citing this case for a proposition about single mentions should carry that objection with it.

Once is enough or once is not enough is not resolved. Claim 6 rests on the dose appearing once, in a range, in a specification about something else. Which of those three features is doing the work, and whether any one alone would suffice, the opinion does not say.

The estoppel error was found and then set aside. The majority treated the district court's judicial-estoppel error as ancillary; the dissent treated it as infecting the whole analysis. This record does not resolve that, and it is the ground the dissent would have remanded on.

Clear-error review does most of the work. Claim 14. Almost every operative statement is that the district court did not clearly err. A different trial record, or a different set of credibility findings, could produce the opposite result under the same law.

What this record does and does not cover

The prosecution-history findings, the written-description standard, the possession analysis, the anchoring argument and the disposition were read end to end, and the dissent was read closely enough to state its two footnoted objections.

What was NOT read closely, and is therefore outside what this record can support: the pharmacology and trial design, the regulatory framework, the full judicial-estoppel argument, and the companion inter partes review. This record says nothing about whether the claims were obvious or anticipated, or about what the Board decided in the parallel proceeding. The district court's findings are known here only through this court's account of them.

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.