Last summer, the Federal Circuit affirmed the Patent Trial and Appeal Board’s determination that claims in a pair of patents owned by Agilent Technologies claiming chemically modified CRISPR guide RNAs are unpatentable. Agilent Techs., Inc. v. Synthego Corp., 139 F.4th 1319 (Fed. Cir. 2025). The court agreed that an earlier published international patent application filed by Pioneer Hi‑Bred sufficiently described these guide RNAs and enabled at least one embodiment within the scope of the challenged claims. The decision highlights the application of the substantial-evidence standard for reviewing the Board’s factual findings, and reinforces the presumption that printed publications are enabled prior art.
Agilent’s claims recite CRISPR guide RNAs that bind to a CRISPR associated protein (“Cas protein”) to form a complex and then direct the formed complex to a specific location on a DNA molecule. Id. at 1322. The Board found that the Pioneer publication defined guide RNAs as RNAs that form a complex with a Cas protein and allow the complexed protein to recognize (and optionally cut) a target DNA sequence. Id. at 1323. Guide RNAs are therefore defined functionally. The Board also determined that Examples 4 and 5 in the Pioneer publication disclosed chemically modified CRISPR guide RNAs having this functionality—even though the examples did not include data demonstrating this functionality. Id. at 1325. Based on these findings, the Board concluded that the Pioneer publication anticipated Agilent’s claims.
On appeal, Agilent argued that the Pioneer publication only outlined a “research plan,” rather than a working disclosure, and therefore did not show the claimed guide-RNA functionality. Id. But the court upheld the Board’s reliance on the publication’s statements that the modified guide RNAs in Examples 4 and 5 form functional Cas-protein complexes that can recognize targeted sites on the DNA molecule. The court deemed these statements substantial evidence of disclosure. Id. at 1325–26. While Agilent asserted the absence of experimental data renders the Pioneer publication non-anticipatory, the court determined that clear assertions of functionality were alone enough disclosure to support anticipation. Id. at 1326.
The court also affirmed the Board’s findings that the Pioneer publication enabled at least one embodiment anticipating the challenged claims. Id. at 1327–30. Agilent contended that CRISPR chemistry in 2014 (when the priority application for the challenged patents was filed) was too undeveloped for such a broad disclosure to be enabling without undue experimentation. But the Board found that nucleic-acid chemistry, CRISPR components, and standard oligonucleotide-synthesis techniques were sufficiently advanced by the relevant date. Id. at 1327–28. The court concluded that this evidence supported the Board’s application of the presumption that a skilled artisan could have made and used at least one of the modified guide RNAs described in the Pioneer publication without undue experimentation. Id. at 1330.
Agilent disagreed with the Federal Circuit’s decision and asked the Supreme Court to revisit the presumption that printed publications are enabled prior art. Agilent argues that this presumption wrongly shifts the burden of proof from the challenger to the patent owner, contrary to 35 U.S.C. §§ 282 and 316(e). Agilent also contends the decision conflicts with the Supreme Court’s 1870 decision in Seymour v. Osborne, 78 U.S. (11 Wall.) 516 (1870), requiring that anticipatory prior art describe a working invention. Agilent believes treating untested (“prophetic”) examples as enabled prior art allows speculative disclosures to invalidate patents that describe actual, working examples.
Agilent’s also challenges the Federal Circuit’s reliance on Rasmusson v. SmithKline Beecham Corp., 413 F.3d 1318, 1326 (Fed. Cir. 2005). Rasmusson’s holding that “proof of efficacy is not required” for anticipation arose in the context of chemical‑compound utility, Agilent notes, not functional biochemistry claims like those requiring guide-RNA functionality. Because Agilent’s claims expressly recite guide RNA functionality, it argues that the absence of any working examples or data demonstrating actual functionality in the Pioneer publication should have foreclosed a finding that the publication includes an enabling disclosure. Taken together, Agilent contends, the court’s decision will make it nearly impossible for patent owners to overcome the presumption of enablement even where the record contains little or no proof of demonstrated functionality.
But the presumption that printed publications are enabled prior art is supported by decades of precedent. See, e.g., Impax Labs., Inc. v. Aventis Pharms., Inc., 545 F.3d 1312 (Fed. Cir. 2008); Amgen Inc. v. Hoechst Marion Roussel, Inc., 314 F.3d 1313 (Fed. Cir. 2003); see also, In re Sasse, 629 F.2d 675 (C.C.P.A. 1980). This precedent consistently treats facially complete technical disclosures as presumptively enabling, placing the burden to produce contrary evidence of non-enablement on patentees or patent applicants. Although the particular facts in those cases differ from Agilent’s situation, those differences may not be enough for the Supreme Court to alter the presumption. Agilent’s arguments about statutory burden might not fully account for the Federal Circuit’s view that §§ 282 and 316(e) concern the ultimate burden to prove unpatentability, not intermediate presumptions about whether a publication qualifies as prior art.
Separately, while Rasmusson highlights an oddity—viz., prior art can anticipate a chemical or biochemical compound even without demonstrating therapeutic use or efficacy—there is a recognized difference between the enablement required of a patent (under 35 U.S.C. § 112) and the narrower enablement required of an anticipatory reference (under 35 U.S.C. § 102). As the court noted, anticipation requires prior art enable only a single embodiment within the claim’s scope. Agilent, 139 F.4th at 1327. Here, the claimed “functionality” is satisfied simply by Cas‑protein binding and DNA targeting, which the Pioneer publication asserts. Id. at 1326–27. And because Agilent’s claims cover chemically modified guide‑RNA configurations, Agilent’s showing that some Pioneer sequences (concerning DNA, not RNA) were nonfunctional did not undermine the Board’s conclusion that the Pioneer publication enabled RNA sequences encompassed by the same claims. Id.
The Federal Circuit’s decision may therefore reflect continuity in the court’s decades old approach to anticipation and printed‑publication prior art. The court seems to have applied a long-standing presumption, deferred to the Board’s factual findings, and distinguished broad disclosures from those enabling at least one operable embodiment. Synthego’s recently filed brief opposing Agilent’s petition presents some of these same points. The Court’s disposition of Agilent’s petition is expected soon.
The presumption of enablement concerns the burden of production and is sensibly applied during prosecution because Patent Office examiners cannot experimentally test the enablement of prior art disclosures, yet may presume as much (a procedural waiver of the examiner’s burden of production). In re Morsa, 713 F.3d 104, 110 (Fed. Cir. 2013). Even in inter partes disputes before the courts or the Patent Office, the challenger is not burdened with proving the prior art enables unless and until the patent owner produces a non-frivolous argument and evidence of non-enablement. Id. The common sense in that procedural exchange makes it unlikely, therefore, that the Supreme Court will eviscerate the presumption’s applicability. In the absence of any change in the law, the Federal Circuit’s decision highlights the evidentiary challenges that patentees and patent applicants face when early, prophetic, but seemingly complete scientific publications surface during prosecution, before the Board, or in litigation.
No doubt the decision has implications beyond the technology at issue. All patent applicants therefore would do well to present a robust claim set understanding that prophetic printed prior art disclosures could later be asserted against the broader claims in the set. Care should also be exercised in preparing a specification capable of supporting claim amendments required to overcome such disclosures, while still protecting the inventive discovery. Proving an anticipatory reference is non‑enabling will remain challenging, given that a disclosure is treated as enabled so long as it can be practiced with routine—though not undue—experimentation. See Enzo Biochem, Inc. v. Calgene, Inc., 188 F.3d 1362, 1371 (Fed. Cir. 1999); see also, In re Wands, 858 F.2d 731, 736–37 (Fed. Cir. 1988).