Can Regenxbio overturn its patent loss against Sarepta Therapeutics?
Zachary Silbersher
Regenxbio previously sued Sarepta Therapeutics for infringement of U.S. Patent No. 10,526,617 (“the ‘617 patent”). The patent is owned by the University of Pennsylvania and exclusively licensed to Regenxbio. Yet, the district court presiding over the lawsuit recently invalidated the ‘617 patent, thereby effectively ending the suit. Regenxbio has already started the appeal process. What is the likelihood Regenxbio can revive its patent claims against Sarepta’s DMD therapies?
Sarepta sells gene-therapy products for treatment of Duchenne muscular dystrophy (“DMD”). As described by the District Court (at 2), Sarepta uses “the [adeno-associated virus] variant rh.74 in cultured host cells to make a gene therapy product referred to as SRP-9001.” (citations omitted). SRP-9001 received FDA approval in 2023 and is sold under the brand name Elevidys®.
Regenxbio’s ’617 patent is directed to methods of detecting and identifying adeno-associated virus (“AAV”) sequences and isolating them. Claim 1 of the patent recites as follows:
1. 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.
During the litigation, Sarepta challenged the patent in two ways. It argued that it did not infringe the patent, i.e., Sarepta claimed its DMD products did not use the invention claimed in the ‘617 patent. Sarepta also challenged the validity of the ‘617 patent. It argued that under Section 101 of the Patent Statute, the claimed invention in the patent was directed to ineligible subject matter, which is another way of arguing that the patent covers something that cannot be patented.
A patent directed to a naturally-occurring substance, by itself, is typically ineligible for patent protection. For instance, a patent directed to the wood that grows on an oak tree would be ineligible for patent protection. Otherwise, the holder of that patent could bar anyone from growing an oak tree or using its wood without paying for a license to that patent. By the same logic, simply combining two naturally-occurring substances could be equally ineligible for a patent protection. A patent combining oak and maple could preclude anyone from building any furniture with those two types of wood without a license. The purpose of Section 101 of the Patent Statute, which defines the subject matter appropriate for patent protection, is therefore intended to stop patents from claiming things that occur in nature, without anything more. Because doing so would impede, rather than encourage, scientific progress.
The court in Regenxbio’s case found that its patent covered naturally-occurring things. The purported “invention” of the ‘617 patent includes a combination of the rh.10 sequence and a heterologous non-AAV sequence. Both of these things appear to be natural products. According to the court, they are essentially a combination of two sequences that have been taken from two different organisms. In other words, they are not manufactured or synthesized. By contrast, the active ingredient in most small-molecule drugs are typically non-naturally occurring synthesized chemicals.
Yet, just because an invention incorporates natural things does not mean it cannot be patented. Indeed, the district court’s opinion in the Regenxbio case identifies patented inventions that incorporate natural things. For instance, if the inventor modified or altered a naturally-occurring substance from how it manifests in nature, that could constitute eligible subject matter. The court cited a prior precedent where the inventor modified a certain bacteria to make it capable of breaking down components of crude oil that the naturally-occurring form of that bacteria could not. Alternatively, a naturally-occurring substance can, in some instances, be patented if the substance is utilized in an inventive way. Yet, in such a case, the patent must “claim” that utility as the invention.
Here, the district court found that Regnenbiox’s patent failed the different ways that naturally-occurring substances can be eligible for a patent. First, Regenxbio argued that it isolated the sequences. That alone, however, is not sufficient. Prior precedents have held that isolated sequences are still naturally-occurring. Moreover, Regenxbio did not purportedly alter or modify the patented sequences in any way. Rather, it combined them. But again, combining two naturally-occurring things still runs afoul of the Patent Statute.
Regenxbio also argued that the patented combination of sequences are useful for gene therapy. Indeed, the ‘617 patent taught that “rAAV vectors of the invention are particularly advantageous in rAAV readministration and repeat gene therapy.” (See ‘617 patent col. 4, line 65-67). Yet, fatal to Regenxbio, the patent did not claim the use of the combined sequences for gene therapy.
This is an important point. Although Regenxbio’s patent may generally teach the use of gene therapy, that use was not included in the patented “claims” that appear at the end of the patent. Those “claims” constitute the actual invention, i.e., the metes-and-bounds of the invention patented by the University. Had the “claims” recited gene therapy as a use for the combined sequences, then the Regenxbio may have prevailed in Sarepta’s invalidity challenge to the patent.
Regenxbio has already started the process of appealing the district court’s decision. On appeal, Regenxbio will have another chance to convince the court that its patent is not directed to naturally-occurring, ineligible subject matter.
The crux of the dispute is whether Regenxbio has modified what occurs in nature sufficiently to be patentable. Regenxbio argues that its patent does not simply claim AAVrh.10 sequences, but “instead cover a markedly different product—cultured host cells that incorporate such sequences along with other components.” (See Case No. 20-cv-1226, Dkt. 221 at 2-3 (D. Del.)). Regenxbio claims this product—this combination of cultured host cells, sequences and other components—simply does not occur in nature. According to Regenxbio, that should be sufficient to pass muster under Section 101 of the Patent Statute.
By contrast, Sarepta argues that the patent simply uses a cultured host cell that contains rh.10 sequences that are naturally-occurring. In other words, using a known container for sequences that naturally occur in nature does not elevate the product to something that is patentable. The issue is likely to be closer on appeal than is evident from the district court’s decision. It’s undoubtedly an abstract question on which reasonable minds, and different judges, could theoretically disagree.
Regardless of the outcome of Regenxbio’s appeal, Sarepta must still defend against other fronts from Regenxbio. The ’617 patent is not the only patent that Regenxbio has asserted against Sarepta. In June 2023, Regenxbio brought a separate litigation against Sarepta that asserts a different patent, U.S. Patent No. 11,680,274 (“the ‘274 patent”). This patent is related to the ‘617 patent and covers similar subject matter. The independent claim of the ‘274 patent recites:
1. A recombinant adeno-associated virus (AAV) comprising an AAV capsid and a minigene having AAV inverted terminal repeats and a heterologous gene operably linked to regulatory sequences which direct expression of the heterologous gene in a host cell, wherein the AAV capsid comprises AAV vp1 proteins, AAV vp2 proteins, and AAV vp3 proteins, wherein the AAV vp1 proteins have i) the sequence of amino acids 1 to 738 of SEQ ID NO: 4 (AAVrh46), or ii) an amino acid sequence at least 95% identical to the full length of amino acids 1 to 738 of SEQ ID NO:4, wherein the amino acid residue corresponding to position 665 in SEQ ID NO: 4 is N when aligned along the full length of amino acids 1 to 738 of SEQ ID NO: 4.
It is not clear on its face whether the ‘274 patent will meet the same fate as the ‘617 patent. The claim appears to also recite certain naturally-occurring sequences, yet whether it actually does so is a question better opined upon by experts within this field of technology.
The claim for the ‘274 patent does not, however, appear to claim any specific utility. It does not appear to claim the utility of gene therapy, which is one of the reasons the ‘617 patent failed its Section 101 challenge. That may prove fateful for the ‘274 patent as it did for the ‘617 patent.
If Regenxbio manages to reverse the invalidity decision on appeal, the case will likely be remanded back to the district court. At that time, Sarepta will likely raise its other primary defense to this litigation—namely, that it purportedly does not infringe the ‘617 patent. Although Serpeta previously moved for summary judgment of non-infringement, the district court denied that motion. That means the Sarepta will be permitted to raise its non-infringement defense again, but the issue will likely have to proceed to trial. Nevertheless, Sarepta has preserved another avenue to escape liability to Regenxbio.