Enablement & Written Description

Enablement

Can a POSITA make and use the full scope of the claims without undue experimentation? Assessed using the 8 Wands factors — breadth of claims matters as much as depth of disclosure.

Written description

Did the inventor actually possess the claimed invention at the filing date? Not whether a POSITA could later figure it out — whether the specification shows the inventor already had. (Ariad, 2010)

Claiming 'all compounds that inhibit X' or 'all antibodies that bind Y' without disclosing specific structural classes that perform the function — written description failure. The specification must show the inventor possessed the full genus, not just the idea of a genus.

Disclosing 2–3 specific compounds and claiming a genus of thousands. The species examples don't demonstrate possession of the genus unless the field is highly predictable. Even in predictable fields, some genus claims fail if there's no unifying structural principle.

Disclosing specific chemical structures, amino acid sequences, genetic sequences, or mechanical structures — and claiming those same structures — satisfies written description even for relatively broad scope.

Disclosing multiple species across a genus with data showing they share the claimed property — especially with a structural explanation for why — provides intermediate support. More species across the range = stronger written description.

What is the difference between enablement and written description under § 112?

Enablement and written description are two separate requirements under 35 U.S.C. § 112(a), confirmed as distinct by the Federal Circuit en banc in Ariad Pharmaceuticals, Inc. v. Eli Lilly and Co. (2010). Enablement asks: can a person of ordinary skill in the art (POSITA) make and use the full scope of the claimed invention without undue experimentation? This is a forward-looking practical question about whether the POSITA could practice the invention based on the specification plus general knowledge in the field. Written description asks: did the inventor actually possess the claimed invention at the time the application was filed? This is a backward-looking possession question — the specification must demonstrate the inventor had actually invented what was claimed, not just that a POSITA could later figure out how to make it. A claim can fail written description even if it is enabled (a POSITA could practice it) if the specification doesn't show the inventor had conceived and possessed it at the filing date. This most commonly arises in genus claims where the inventor discloses a few specific species but claims a broad functional genus without showing they possessed the full genus.

What are the Wands factors for patent enablement?

The Wands factors come from In re Wands (Federal Circuit 1988) and are the framework courts and examiners use to determine whether a patent specification enables the claimed invention without 'undue experimentation.' The eight Wands factors are: (1) quantity of experimentation required to practice the claims; (2) amount of direction or guidance provided in the specification; (3) presence or absence of working examples; (4) nature of the invention (simple or complex field); (5) state of the prior art; (6) relative skill of those in the art; (7) predictability or unpredictability of the art; and (8) breadth of the claims. No single factor is dispositive — they are weighed together. Crucially, 'undue experimentation' does not mean no experimentation is permitted. A POSITA can perform routine experimentation based on specification guidance. What is impermissible is requiring the POSITA to engage in extensive, creative, or research-level experimentation to practice the claims.

What did Amgen v. Sanofi hold about enablement for antibody patents?

Amgen Inc. v. Sanofi (Supreme Court 2023) was a landmark enablement case in which the Supreme Court unanimously invalidated Amgen's broad PCSK9 antibody claims for failure to meet the enablement requirement of § 112. Amgen claimed a genus of antibodies that bind to PCSK9 at specific sites and block it from binding to LDL receptors — a functionally defined genus that encompassed potentially millions of antibody variants. Amgen had identified only about 26 specific antibodies in its specification. The Court held that the patent failed enablement because Amgen provided only a roadmap (an amino acid sequence to immunize and test) for discovering the remaining antibodies in the claimed genus — the POSITA would have had to engage in substantial trial-and-error creative research to make and use the full scope of the claim. The decision reaffirmed a key principle: to claim a broad genus, you must enable the full scope of the genus — not just disclose a process for discovering members of the genus. Amgen has significant implications for pharmaceutical and biotechnology patent drafting, especially for functional genus claims (antibodies defined by what they bind or block rather than by structural features).

How should patent specifications be written to satisfy § 112(a)?

Best practices for satisfying § 112(a) at drafting: (1) Match claim breadth to specification depth. Don't draft broad genus claims without broad enabling disclosure. If your claims cover all antibodies that inhibit function X, your specification should describe or demonstrate how to make many of them across the range — not just a single example. (2) Include working examples. Real experimental data is stronger than prophetic examples. In pharma/biotech, actual synthesis and testing data is often essential. (3) Describe structural features for functional claims. Functional genus claims ('all antibodies that bind to X') are most vulnerable — add structural disclosure showing specific structural classes that perform the function, with data demonstrating the structural-functional relationship. (4) Explicitly describe each embodiment within the claimed scope. For ranges, provide data at multiple points in the range. For species/genus relationships, provide representative species across the genus. (5) At filing, don't narrow claims faster than the disclosure supports. Claims can always be narrowed in prosecution; the specification cannot be changed to add new enablement support after filing. (6) For CIP applications, ensure any new matter added is itself adequately enabling — the original parent's disclosure cannot retroactively enable new CIP claims.

Can an issued patent be invalidated for lack of enablement?

Yes — lack of enablement and lack of written description are grounds for invalidity under 35 U.S.C. § 282, and either can be raised in district court litigation or in post-grant proceedings at the PTAB. In district court: invalidity must be proven by clear and convincing evidence (the patent carries a presumption of validity). In PTAB inter partes review (IPR): only § 102 and § 103 grounds are available — § 112 grounds cannot be asserted in IPR. In post-grant review (PGR): § 112 grounds ARE available in PGR, but only for patents with an effective filing date of March 16, 2013 or later (AIA first-inventor-to-file patents), and only if the PGR petition is filed within 9 months of the patent grant. Ex parte reexamination: § 112 cannot be raised as a new ground of rejection in reexamination unless the examiner independently identifies the issue. Enablement and written description are particularly powerful invalidity arguments in biotech and pharmaceutical patent litigation, where broad functional claiming is common and the Amgen v. Sanofi decision has weakened many genus claims.