Plot Twist: Splicing Together DNA From Two Species Is NOT The Same As Mixing Yogurt Cultures
Friends, colleagues, fellow Section 101 trauma survivors—gather ‘round, because Judge Kara Stoll just dropped what I'm calling the tour de force of patent eligibility in biopharma with REGENXBIO v. Sarepta Therapeutics. [1] Pour one out for the district court’s Funk Brothers [2] analogy, because the Federal Circuit unanimously said “that's gonna be a ‘no’ from us, dawg.”
On February 20, 2026, the U.S. Court of Appeals for the Federal Circuit reversed a Delaware district court’s summary judgment of ineligibility, holding that REGENXBIO's gene therapy patent claims are not directed to a natural phenomenon. In doing so, Judge Stoll—joined by Judges Dyk and Hughes—provided a masterclass in how to properly apply the “markedly different characteristics” test from Diamond v. Chakrabarty [3] to modern biotechnology. And the biopharma world is collectively exhaling.
The Setup: Gene Therapy, Billion-Dollar Drugs, and a $900 Million Lawsuit
Let's set the stage. REGENXBIO and the Trustees of the University of Pennsylvania sued Sarepta Therapeutics over U.S. Patent No. 10,526,617 (“the ‘617 patent”), which covers genetically engineered host cells containing adeno-associated virus (AAV) rh.10 sequences. The technology at issue is foundational to gene therapy—a field that treats genetic disorders like cystic fibrosis, hemophilia, sickle cell anemia, and Duchenne muscular dystrophy by using modified virus vectors to deliver therapeutic genes that replace defective or missing ones.
This isn't academic. Sarepta’s Elevidys—a gene therapy for Duchenne muscular dystrophy that costs more than most houses (we're talking $3.2 million per dose) [4]—was approved by the FDA in 2023 and generated $898.7 million in revenue in 2025. [5] REGENXBIO has sought more than $900 million in damages, alleging Sarepta used the patented AAV platform without authorization.
Representative claim 1 of the '617 patent recites:
“A cultured host cell containing a recombinant nucleic acid molecule encoding an AAV vp1 capsid protein having a sequence comprising amino acids 1 to 738 of SEQ ID NO: 81 (AAVrh.10) or a sequence at least 95% identical to the full length of amino acids 1 to 738 of SEQ ID NO: 81, wherein the recombinant nucleic acid molecule further comprises a heterologous non-AAV sequence.”
Translation for the non-patent crowd: this claims a lab-created cell containing an engineered DNA molecule that splices together bits from different species to produce “AAV capsid proteins” useful for gene delivery. The cells are undisputedly human-made. They do not exist in nature.
The District Court's Ruling: “It’s Just Like Mixing Bacteria Together”
In January 2024, the Delaware district court granted summary judgment for Sarepta, holding the claims patent-ineligible under 35 U.S.C. § 101. [6] The district court's reasoning? Taking "two sequences from two different organisms and putting them together" is "no different than taking two strains of bacteria and mixing them together"—citing the Supreme Court's 1948 decision in Funk Brothers Seed Co. v. Kalo Inoculant Co.
I'm sorry, WHAT?
The district court further reasoned that none of the individual naturally occurring components had been "changed," and that simply combining natural products in a host cell doesn't make an invention patentable. The court also found no “inventive concept” under the second step of the Alice/Mayo framework because “the claimed invention is made using well-understood, routine, and conventional steps.”
As Kevin Noonan over at Patent Docs put it, after a decade of the patent defense bar pushing ever more extreme positions that made Section 101 invalidation the default outcome, the envelope had finally been pushed too far. [7]
The Federal Circuit's Response: “That’s...Not How Any of This Works.”
The Federal Circuit unanimously reversed, and Judge Stoll's opinion reads like a patient but firm correction of a student who clearly didn't do the reading.
The court began by explaining that Chakrabarty defines the inquiry: does the claimed composition have “markedly different characteristics” and "the potential for significant utility" compared to what naturally occurs? Applying this framework, the Federal Circuit concluded that the claims here are more analogous to the eligible claims in Chakrabarty and the cDNA claims in Myriad than to the ineligible claims in Funk Brothers, Myriad's isolated gene claims, or ChromaDex.
Here's the key passage:
“It is uncontested that the claimed host cells include a recombinant nucleic acid molecule that does not and cannot exist in nature. Specifically, the claims require (1) ‘recombinant’ nucleic acid, which means segments of nucleic acid from one source are artificially manipulated or inserted into the nucleic acid of another source through gene splicing; and (2) a nucleic acid molecule capable of encoding a sequence at least 95% identical to an AAV rh.10 sequence and a ‘heterologous’ non-AAV sequence, where ‘heterologous’ means from a different species. Thus, the recombinant nucleic acid molecule must be spliced together via human intervention from at least two different species to meet the claim limitations.”
The court emphasized that the claimed nucleic acid molecules—although containing naturally occurring segments of DNA—are “not nature’s handiwork” but rather “a nonnaturally occurring manufacture or composition of matter.”
Why Funk Brothers Doesn't Apply: It’s Not a Yogurt Parfait, Folks
The Federal Circuit found the district court's Funk Brothers analogy “flawed and inconsistent with the undisputed scientific evidence in the record.”
In Funk Brothers, the patentee had discovered that certain root-nodule bacteria didn't inhibit each other when mixed together, and created a commercial product packaging multiple bacteria strains. The Supreme Court found this ineligible because the bacteria "perform in their natural way" and the combination "does not improve in any way their natural functioning." It was, at best, a packaging improvement.
The claims here are fundamentally different. As the Federal Circuit explained:
“Genetically engineering two nucleic acid sequences from separate species into a single molecule and then transforming a host cell in order to incorporate that new molecule into it—thus fundamentally creating a cell containing a molecule that could not form in nature on its own—is materially different from growing more than one naturally occurring bacteria strain in a culture where none of the bacteria undergo any change from their natural state.”
Creating recombinant DNA is a multi-step laboratory process involving restriction enzymes to cleave DNA at specific sites, DNA ligases to join fragments, ligation into plasmids, transfection to artificially introduce the nucleic acids into cells, and amplification through host cell colonies. As amici from the Parker Institute for Cancer Immunotherapy, The J. David Gladstone Institutes, and the Dana-Farber Cancer Institute emphasized, this is sophisticated human engineering—not tossing ingredients in a jar together.
The Key Precedents: A Tour Through Patent Eligibility’s Greatest Hits
Judge Stoll’s opinion provides a comprehensive walkthrough of the governing Supreme Court precedents:
Diamond v. Chakrabarty (1980): The Court held eligible claims to a genetically engineered bacterium capable of breaking down crude oil. The claimed bacterium was “a nonnaturally occurring manufacture or composition of matter—a product of human ingenuity having a distinctive name, character and use.” Importantly, the Federal Circuit noted that Chakrabarty’s inventors transferred four naturally occurring plasmids into a bacterium—so the individual components weren't themselves changed—yet the resulting composition was still patent-eligible because the whole was markedly different from anything in nature.
Funk Brothers Seed Co. v. Kalo Inoculant Co. (1948): The Court found ineligible a mixed culture of root-nodule bacteria because “each species has the same effect it always had” and “the bacteria perform in their natural way.” This was mere packaging, not transformation.
Association for Molecular Pathology v. Myriad Genetics, Inc. (2013): The Court held that isolated DNA encoding BRCA1 and BRCA2 genes was ineligible because Myriad “did not create anything”—the location and order of nucleotides existed in nature. But the Court found cDNA (complementary DNA) patent-eligible because “the lab technician unquestionably creates something new when cDNA is made.”
ChromaDex, Inc. v. Elysium Health, Inc. (Fed. Cir. 2023): The Federal Circuit found claims to isolated vitamin NR from milk ineligible because “the asserted claims do not have characteristics markedly different from milk.”
The Federal Circuit distinguished the present claims from the ineligible claims in each of these cases, emphasizing that the claimed host cells here contain recombinant nucleic acid molecules that "by definition" are markedly different from anything occurring in nature.
Don’t Dissect the Claims: Diamond v. Diehr Says Hi
One of the most important aspects of the opinion is the Federal Circuit's rejection of Sarepta's attempt to focus only on the AAV rh.10 sequence while dismissing other claim elements as “conventional.”
Citing Diamond v. Diehr, 450 U.S. 175, 188 (1981), where the Court stated: “[i]t [is] inappropriate to dissect the claims into old and new elements and then to ignore the presence of the old elements in the [§ 101] analysis,” the Federal Circuit declined to “read out or ignore limitations in a claim merely because they may be found in the prior art or within the knowledge of a skilled artisan.”
This is a critical point. As the court noted in a footnote, while this may be material for novelty and obviousness inquiries under §§ 102 and 103, parties and tribunals should not conflate the separate novelty and obviousness inquiries with the step one inquiry under § 101.”
Industry Reaction: Relief, Vindication, and a Collective Sigh
The reaction from the biotech community has been swift and overwhelmingly positive. And the stakes for the industry couldn’t be higher. According to Research and Markets, the AAV gene therapy market reached $1.9 billion in 2023 and is forecasted to exceed $75 billion by 2034. [8] Major amici had lined up to support REGENXBIO's appeal, including the Parker Institute for Cancer Immunotherapy, The J. David Gladstone Institutes, Dana-Farber Cancer Institute, former Federal Circuit Chief Judge Paul R. Michel, and the American Intellectual Property Law Association.
Key Takeaways for Patent Drafters and Life Sciences Companies
The decision offers several practical lessons:
Drafting matters—a lot. Claim terms like “recombinant” and “heterologous” proved outcome-determinative. These terms anchor the invention to a human-made construct that cannot exist in nature. Patent applications should emphasize human manipulation and marked differences from natural products.
Evaluate the composition as a whole. Courts should not focus narrowly on whether individual components are naturally occurring. The question is whether the claimed composition as a whole is “not naturally occurring.”
Utility can be implicit. The court may consider whether the claimed composition has “the potential for significant utility” even if that utility isn't expressly recited in the claims. Here, the specification's disclosure that embodiments “are beneficial for gene delivery to selected host cells and gene therapy patients” supported eligibility.
Don't conflate § 101 with §§ 102/103/112. Claim breadth concerns should be addressed through novelty, obviousness, and enablement—not eligibility. Section 101 is meant to be a low bar.
Funk Brothers has limits. Courts may resist product-of-nature framing where the claim requires engineered combinations that cannot arise naturally. There's a fundamental difference between packaging natural materials together and constructing something new through genetic engineering.
The Bottom Line: If Nature Can’t Make It, Maybe It's Not a “Product of Nature”
If your invention requires a lab technician to create something that nature could never assemble on its own—using restriction enzymes to cleave DNA at specific sites, ligases to join fragments, transfection to artificially introduce nucleic acids into cells—maybe, just maybe, it's not a “product of nature.”
The Federal Circuit has now put that principle in writing, and Judge Stoll's opinion should be required reading for every examiner who's ever rejected a biotech claim under § 101 by focusing on whether the individual nucleotides themselves exist in the wild.
Congratulations to REGENXBIO, UPenn, and their counsel at Fish & Richardson and Morgan Lewis on the reversal. Now back to the district court you go—with your patent intact—to litigate infringement, damages, and whatever other validity challenges Sarepta wants to raise on grounds where they actually belong.
After nearly 100 cert denials on Section 101 issues, it seems the Federal Circuit is ready to be the adult in the room. And for the gene therapy industry, that's very good news indeed.
This publication and/or any linked publications herein do not constitute legal, accounting, or other professional advice or opinions on specific facts or matters and, accordingly, the author(s) and PierFerd assume no liability whatsoever in connection with its use. Pursuant to applicable rules of professional conduct, this publication may constitute Attorney Advertising. © 2026 Pierson Ferdinand LLP.
[1] REGENXBIO Inc. v. Sarepta Therapeutics, Inc., No. 24-1408 (Fed. Cir. Feb. 20, 2026).
[2] Funk Brothers Seed Co. v. Kalo Inoculant Co., 333 U.S. 127 (1948).
[3] Diamond v. Chakrabarty, 447 U.S. 303 (1980).
[4] https://www.biopharmadive.com/news/sarepta-duchenne-elevidys-price-million-gene-therapy/653720/
[6] REGENXBIO Inc. v. Sarepta Therapeutics, Inc., No. 20-cv-1226-RGA, 2024 WL 68278 (D. Del. Jan. 5, 2024).
[7] https://patentdocs.org/2026/02/23/regenexbio-inc-v-sarepta-therapeutics-inc-fed-cir-2026/
[8] https://www.researchandmarkets.com/reports/5969642/adeno-associated-virus-gene-therapy-market
