In a significant § 101 decision for biotech and gene therapy, the Federal Circuit reversed the District of Delaware’s summary‑judgment ruling that the claims of U.S. Patent No. 10,526,617 (“the ’617 patent”) were ineligible under § 101. The decision clarifies that recombinant, human‑engineered host‑cell compositions—viewed as a whole—remain patent‑eligible, even when their constituent sequences have natural origins. REGENXBIO Inc. v. Sarepta Therapeutics, Inc., No. 2024‑1408 (Fed. Cir. Feb. 20, 2026) (Stoll, J.).
REGENXBIO et al.—owners of the ’617 patent—initially sued Sarepta in the District of Delaware for patent infringement based on Sarepta’s alleged use of the patented cultured host cells to manufacture SRP‑9001, a gene‑therapy candidate for Duchenne muscular dystrophy. In its first significant move, Sarepta sought early dismissal under the § 271(e)(1) safe harbor, arguing that its use of the patented host cells was “reasonably related” to preparing regulatory submissions for the FDA. The safe harbor, enacted through the Hatch‑Waxman Act, shields certain otherwise infringing acts when performed solely to generate information for FDA review.
The District of Delaware rejected that argument. Applying Proveris, the court held that because the patented cultured host cells are not themselves subject to FDA premarket approval, they cannot be a “patented invention” within the meaning of § 271(e)(1). As a result, Sarepta could not invoke the safe harbor at the pleading stage, and the infringement suit proceeded to summary judgment—where the § 101 controversy ultimately emerged.
What the Federal Circuit Decided—And Why it Matters
On summary judgment, the district court held that the asserted claims of the ’617 patent were ineligible because both the AAVrh.10 capsid sequence and the heterologous non‑AAV sequence were “natural,” and their combination in a host cell did not confer patent‑eligible subject matter. Relying on Funk Bros. Seed Co. v. Kalo Inoculant Co., the court analogized the claimed invention to a mere aggregation of natural components behaving no differently together than in isolation. For contextual grounding, adeno‑associated viruses (AAVs) are naturally occurring viral vectors widely used in gene‑therapy manufacturing, precisely because they can deliver genetic material into human cells.
The Federal Circuit rejected this component‑by‑component approach as both scientifically flawed and inconsistent with § 101 jurisprudence. The proper question, the panel emphasized, is whether the claimed composition possesses “markedly different characteristics” from anything in nature—not whether each ingredient has a natural origin or has been chemically altered. Applying Diamond v. Chakrabarty, 447 U.S. 303 (1980), and Assoc. for Molecular Pathology v. Myriad Genetics, Inc., 569 U.S. 576 (2013), the court explained that a recombinant nucleic acid assembled from sequences drawn from different species, and then introduced into a cultured host cell, “does not and cannot exist in nature.” Consistent with Chakrabarty and Myriad, the panel reaffirmed that the laboratory technician “unquestionably creates something new.”
Claim 1 of the ’617 patent recites “a cultured host cell containing a recombinant nucleic acid molecule” encoding an AAV vp1 capsid (AAVrh.10; SEQ ID NO. 81) or a sequence at least 95% identical to that sequence, and a “heterologous non‑AAV sequence.” The Federal Circuit highlighted several definitional anchors:
- The cultured host cells are undisputedly human‑made;
- “Recombinant” requires artificial gene splicing across biological sources; and
- “Heterologous” means originating from a different species.
These features drove the conclusion that the claimed cultured host cells possess structural and functional attributes “markedly different” from anything found in nature.
Why Funk Bros. Didn’t Apply
The panel distinguished Funk Bros. Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948), as a classic “repackaging” case—combining intact bacteria that behaved no differently together than apart. By contrast, the ’617 patent claims a single recombinant construct, created by splicing sequences from different organisms and transforming a host cell to incorporate that engineered molecule. The district court’s analogy—treating “two sequences from two organisms” as equivalent to “two strains in a bag”—misapprehended the transformative nature of recombinant gene assembly.
Alice/Mayo—And What the Panel Didn’t Need to Do
Echoing its reasoning in ChromaDex, Inc. v. Elysium Health, Inc., 59 F.4th 1280 (Fed. Cir. 2023), the Federal Circuit explained that its analysis could have ended under Chakrabarty’s markedly‑different‑characteristics test alone. Even if the Alice/Mayo framework were applied, the claims readily pass step one for the same reasons—rendering step two unnecessary. The panel also reiterated that conventionality and novelty should not be imported into § 101; those questions belong under §§ 102 and 103.
Strategic Takeaways
Biotech and gene therapy patentees and applicants should draft composition claims that require recombinant nucleic acids assembled from different biological sources and housed in engineered host cells. Because the composition is non‑natural, such claims are well‑positioned to survive § 101 challenges—even when their components originate in nature.
Litigators should frame § 101 arguments around whole‑composition analysis and the human‑made recombinant assembly. Push back against attempts to strip out “conventional” limitations at step one; Diehr bars dissecting claims into old versus new elements for § 101 purposes.
As for investors, the decision narrows patent‑eligibility risk for manufacturing‑stage cell compositions and AAV platform assets—key considerations in diligence and valuation of gene‑therapy portfolios.
Bottom Line
REGENXBIO v. Sarepta reinforces that engineered biological compositions—particularly those involving cross‑species recombinant constructs—are patent‑eligible under § 101. When human ingenuity recombines genetic material across species into a single engineered molecule and installs it in a cultured host cell, the resulting composition is not nature’s handiwork. This decision will likely feature prominently in upcoming life‑sciences eligibility disputes. If you have any questions, feel free to reach out to your Dinsmore attorney.