Most are familiar with the requirement that a patent application must contain an adequate written description to justify the application’s claims and invention enablement under 35 U.S.C. § 112(a). These two requirements for patentability are in addition to the requirement that the application’s claims are not anticipated or obvious over the prior art. The written description and enablement requirements also apply to any previously filed application for which the application undergoing examination is claiming priority. Written description issues/enablement issues may also arise when a patent’s validity is challenged.
This blog discusses the Court of Appeals of the Federal Circuit’s (“the CAFC”) December 2025 decision in Seagen Inc. and Daiichi Sankyo, et al. where it held that Seagren’s patent could not rely on the filing date of an earlier-filed patent application as its priority filing date because the earlier-filed patent application lacked an acceptable written description. Nor was Seagen’s patent enabling. As such, Seagen’s patent was declared invalid because Daiichi’s disclosure in relevant prior art predated Seagren’s patent application’s actual filing date.
The Seagen decision is instructive for its cogent discussion of the two important requirements for establishing a patent application’s effective filing date: 1) for a patent to “gain the benefit” of an earlier application’s priority date, the earlier application must have disclosed an invention in a written description that meets the requirements of 35 U.S.C. § 112(a); and 2) the earlier application must be enabling under 35 U.S.C. § 112(a) meaning that the claimed invention must be capable of being made and used without undue experimentation by a skilled artisan in the relevant art.
Background
Seagen’s U.S. patent 10,808,039 (“the ’039 patent”) is for a cancer therapeutic claimed to specifically target cancer cells while sparing healthy cells. Seagen sued Daiichi for patent infringement. The district court jury awarded Seagen $41 million in damages. The district court denied Daiichi’s post-trial motion for judgment as a matter law (JMOL) that Seagen’s patent lacked a written description and failed to meet the enable requirement. Daiichi appealed to the CAFC.
Dates turned out to be important in the CAFC’s decision. Seagren had filed an application in 2004. Apparently, the application never evolved into a patent. Seagren then filed a new application in 2019 claiming priority to the 2004 application.
Daiichi developed a therapeutic in 2011 and disclosed its structure and functionality publicly in December 2015. Tellingly, the CAFC opined that Seagen’s ’039 patent application seemed to have been “filed to specifically encompass Daiichi’s invention, which of course is permissible if it was entitled to a filing date antedating the public disclosure of Daiichi’s invention.” Slip opinion at 6.
The Written Description Issue
In order for a later-filed patent application to “gain the benefit” of an earlier application’s filing date, the earlier application’s written description must adequately describe the invention. This means that the written description must “reasonably convey to those skilled in the art that the inventor had possession of the claimed subject matter as of the filing date.” Here, the relevant priority date was deemed to be in 2004. Slip opinion at 10.
The ’039 patent, claiming priority to the 2004 application, disclosed an antibody-drug conjugate (“ADC”) comprised of an antibody modified to include a tetrapeptide (i.e., a sequence of 4 amino acids) separated from the cancer-therapeutic drug moiety by a spacer unit. Only 2 amino acids were involved – glycine and phenylalanine wherein the tetrapeptide must include both of the amino acids. Phenylalanine, as the result of its structure, can exist in two 3D spatial arrangements while glycine only has one spatial arrangement. Based on statistics, Seagen’s ’039 patent thus covered 34 (81) potential different species of the disclosed tetrapeptide chain.
So far so good. Unfortunately for Seagen, its 2004 patent application did not specifically outline possible structures of the peptide unit it claimed in the 2019 application. For example, the 2004 application stated that the peptide unit could range from 2-12 amino acids. Then, this application described 39 different amino acids (totaling 83 unique options based on the amino acids different isomers) which could comprise the peptide unit. Based on statistical analysis, the possibilities were 47 million combinations! The commentator notes that there are only 22 amino acids routinely used by the body for protein synthesis. Accordingly, that earlier application included 17 uncommon amino acids. Glycine and phenylalanine are in the group of 22.
The technical question was: which of the possible combinations in the 2004 application would yield the desired pharmaceutical effect? The CAFC concluded that the answer to this technical question is, who knows?
The CAFC held that Seagen’s 2004 application’s broad disclosure of an extraordinary number of peptide units did not provide a “reasonably specific supporting disclosure” for the claimed subgenus in the 2019 application’s 83 Gly/Phe-only tetrapeptides. Slip opinion at 12. Even the 2004 application’s inventor stated in district court testimony that the “first time . . . [he] ever saw a Gly/Phe-only tetrapeptide in the claimed therapeutic was in Daiichi’s Enhertu formulation. As the CAFC stated, “because one cannot describe what one has not conceived,” the 2004 inventor’s admissions supported Daiichi’s assertion that they did not possess the claimed subgenus Gly/Phe tetrapeptides as of the November 2004 priority date.” Slip opinion at 13.
As such, Seagen could not rely on the 2004 patent application to predate Daiichi’s Daiichi’s 2015 disclosure because the 2004 patent application lacked a written description sufficient for showing that the inventors actually “had possession” of the invention disclosed in the ’039 patent in 2004. The applicable prior art date in Seagen’s infringement action against Daiichi was thus any publication before the filing date of Seagen’s 2019 application, in this case Daiichi’s 2015 disclosure. The CAFC concluded that the district court erred in denying Daiichi’s JMOL motion that Seagen’s patent failed to meet the written description requirements. “The testimony [from the district court trial] does not show that the 2004 application provides ‘reasonably specific supporting disclosure” for the claimed Gly-Phe-only tetrapeptides or that the inventors had possession of that subgenus of tetrapeptides” in 2004. Slip opinion at 14.
The Enablement Issue
The enablement requirement under 35 U.S.C. § 112(a) is distinct from the written description requirement although the two requirements are often confused. Citing precedent, the CAFC stated, “[to] be enabling, the specification of a patent must teach those skilled in the art how to make and use the full scope of the claimed invention without undue experimentation.” Slip opinion at 16. It’s important to emphasize that to be enabling, any resulting attempt to make and use of the invention need not result in perfection or that the invention even be commercially viable. Section 2164.01(a) of the MPEP lists the following factors that are used to determine whether or not an invention is enabling:
(A) The breadth of the claims;
(B) The nature of the invention;
(C) The state of the prior art;
(D) The level of one of ordinary skill;
(E) The level of predictability in the art;
(F) The amount of direction provided by the inventor;
(G) The existence of working examples; and
(H) The quantity of experimentation needed to make or use the invention based on the content of the disclosure.[1]
Claim 1 of Seagen’s ’039 patent recited the functional limitation: “wherein the drug moiety is intracellularly cleaved in a patient from the antibody of the antibody-drug conjugate or an intracellular metabolite of the antibody-drug conjugate.” Slip opinion at 5-6. The CAFC found that the enablement requirement was not met because neither the 2004 application’s nor the ’039 patent’s written description discussed a quality common to every functional embodiment of the ADC that would fairly accurately predict such an outcome in all cases. Rather, the CAFC emphasized, “the science is so unpredictable that a skilled artisan would be required to use assays to test whether any given drug moiety meets this functional limitation.” That is, whether the drug is actually cleaved from the ADC within the target type of cancer cell would need to be evaluated in vitro. As the CAFC stated, “the amount of trial-and-error discovery that would be required to determine whether an ADC with a given drug intracellularly cleaves is further increased by the need to test each ADC with the recited structural limitation.” Slip opinion at 18. As such, the ’039 patent failed to meet the enablement requirement.
As for any particular ADC’s performance as a cancer-fighting pharmaceutical upon of the drug moiety, the commentator adds that even encouraging in vitro test results do not necessarily translate into acceptable in vivo (i.e. patient) results. Pharmaceutical biotech inventions are therefore subject to a high rate of commercialization failure. Any such medical treatment also requires FDA approval based on clinical trial data. Daiichi has received FDA approval for Enhertu® for the treatment of certain cancers using its ADC demonstrating that it was able to reduce its “Gly-Phy” ADC to actual useful practice as a cancer therapeutic.
Discussion
The CAFC’s conclusion that Seagen did not have possession of the invention in 2004 given the humungous combination of antibody-conjugate possibilities is not surprising based on the CAFC’s precedent. Even with the drastic reduction to only the 83 combinations of the ’039 patent (compared to the millions of possibilities in the 2004 application), the unpredictability of the art would still require a lot of undue experimentation to verify that any possible combination of Gly/Phe in the tetrapeptide would work with different chemotherapy drugs.
The commentator notes that the ’039 patent also claims a spacer; the impact of this spacer on the purported functionality was not addressed in the decision. Quite apart from the tetrapeptide chain, its amino acid composition, and the spacer composition, the distance between the antibody moiety and the drug may also affect how the cancer therapeutic is metabolized in the body (preferably in a cancer cell only). The commentator has a background in developing antibody markers in the medical diagnostics arena which required a lot of experimentation to generate the “best” antibody reagent for evaluating e.g., a blood protein indicative of certain health problem. However, the number of potential variants did not begin to approach 81.
In the Seagen case, getting it right meant developing a treatment that provides statistically significant benefits without causing undue harm to the patient. All of the claimed permutations could not be expected to perform with the same degree of efficacy even if the ADCs actually underwent the predicted cleavage in a cancer cell to release the drug agent.
It must be emphasized that the examining attorney who granted the ’039 patent apparently did not question the written disclosure or find that the invention was not enabled. Further, the jury had found that Seagen’s 2004 application met both the written description requirement and enablement requirements so as to allow it to serve as the ’039 patent’s effective filing date. The Seagen decision drives home the point that a patent can encounter validity problems long after the patent was granted. In the commentator’s opinion the long time between the 15-year window separating Seagen’s 2004 application from the 2019 filed application did not help its cause.
Take Home Points
An applicant who plans to rely on an earlier filed application as a later application’s effective filing date for prior art disqualification reasons needs to ensure that the earlier filed application has an adequate written description and is enabling. For example, provisional patent applications (PPAs) are routinely filed to get the all-important filing date because any forthcoming patent will issue to the first inventor to file absent unusual circumstances. If a utility patent application (UPA) is subsequently filed within 1 year of the filing date of a PPA, the UPA will generally claim priority to the PPA to try to “push back” any prior art references to dates before the PPA’s filing date. To take advantage of this scenario, PPA filers should ensure that their PPA has an adequate written disclosure for supporting future UPA’s subject matter and claims and enablement of the invention. As we saw in Seagen, failure to do so will allow the examiner and the courts to rely on prior art existing up to the UPA’s filing date, thereby opening up the possibility of more problematic prior art for use in rejecting claims.
[1] Amgen v. Sanofi, 598 U.S. 594 (2023).
Susan Dierenfeldt-Troy
