Sarepta Therapeutics Inc v. Trustees Of University Of Pennsylvania

1. Case Identification

• Case #: IPR2024-00580
• Patent #: 11,680,274
• Filed: February 21, 2024
• Petitioner(s): Sarepta Therapeutics, Inc.
• Patent Owner(s): The Trustees of the University of Pennsylvania and REGENXBIO Inc.
• Challenged Claims: 1, 3-6, and 8

2. Patent Overview

• Title: Method of Increasing the Function of an AAV Vector
• Brief Description: The ’274 patent relates to recombinant adeno-associated virus (rAAV) vectors for gene therapy. The claims are directed to an rAAV with a capsid protein sequence that is at least 95% identical to the AAVrh.46 variant and specifically requires the amino acid asparagine (N) at position 665.

3. Grounds for Unpatentability

Ground 1: Claims 1 and 3-6 are obvious over the ’772 Publication

• Prior Art Relied Upon: The ’772 Publication (Application # 2003/0138772).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued the ’772 publication discloses all elements of the challenged claims. Specifically, it discloses a preferred rAAV embodiment, AAVrh.10, with a vp1 capsid protein sequence that is 97.29% identical to the claimed AAVrh.46 sequence, meeting the limitations of claims 1 and 3. While AAVrh.10 has a serine (S) at position 665, the ’772 publication also discloses another preferred embodiment, AAV8, which has the claimed asparagine (N) at that same position. The publication further discloses the remaining elements of dependent claims 4-6, including the use of heterologous inverted terminal repeats (ITRs), physiologically compatible carriers, and a list of potential therapeutic genes.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSA) would be motivated to create an artificial vector combining the properties of the two preferred embodiments disclosed and compared in the ’772 publication. A POSA would seek to impart the superior liver transduction of AAV8 onto the AAVrh.10 vector (noted for its "superb" lung tropism) by making substitutions. The publication explicitly teaches creating "artificial AAV serotypes" by combining sequences from different variants to confer favorable properties.
  • Expectation of Success: A POSA would have a reasonable expectation of success in making a single amino acid substitution (S665N) in AAVrh.10 to create a viable and potentially superior vector. Such single-point mutations were routine, and Petitioner noted that the Patent Owner’s expert in a related litigation confirmed that the ’772 publication teaches the creation of such artificial variants.

Ground 2: Claims 1 and 3-6 are obvious over the ’772 Publication in view of Xie

• Prior Art Relied Upon: The ’772 Publication (Application # 2003/0138772) and Xie (a 2002 journal article on AAV2 crystal structure).
• Core Argument for this Ground:
  • Prior Art Mapping: The core teachings from the ’772 publication are the same as in Ground 1. Xie provides the three-dimensional crystal structure of the AAV capsid and teaches that amino acids on the capsid surface are critical for governing viral characteristics like tropism, transduction efficiency, and immunogenicity.
  • Motivation to Combine: A POSA, motivated by the ’772 publication to combine features of AAVrh.10 and AAV8, would consult known structural data like that in Xie to identify the most promising amino acid positions for modification. Xie teaches focusing on surface-exposed residues. Of the 48 amino acids that differ between AAVrh.10 and AAV8, Xie would direct a POSA to the 18 that lie on the capsid surface, including position 665. This would provide a strong rationale for prioritizing the S665N mutation.
  • Expectation of Success: Because Xie identifies surface residues as being located in more flexible loop structures that are more tolerant of substitution, a POSA would have a high expectation of success that modifying the surface-exposed serine at position 665 would result in a stable, functional capsid with altered biological properties.

Ground 3: Claims 1 and 3-6 are obvious over the ’772 Publication in view of Snowdy

• Prior Art Relied Upon: The ’772 Publication (Application # 2003/0138772) and Snowdy (a 2003 doctoral thesis).
• Core Argument for this Ground:
  • Prior Art Mapping: Snowdy teaches a method for improving AAV transduction efficiency by mutating phosphorylatable amino acids (e.g., serine) to non-phosphorylatable ones (e.g., asparagine). The serine (S) at position 665 of AAVrh.10 is a predicted phosphorylation site, whereas the asparagine (N) at the corresponding position in AAV8 is non-phosphorylatable.
  • Motivation to Combine: Snowdy provides an independent motivation for a POSA to make the specific S665N mutation in AAVrh.10. A POSA seeking to improve the transduction efficiency of the AAVrh.10 vector would apply Snowdy’s teachings and identify the S665N substitution (which mirrors the sequence of AAV8) as one of only four promising mutations for removing a phosphorylation site.
  • Expectation of Success: Snowdy reported that all tested mutants involving the removal of a phosphorylation site were capable of forming virus at normal titers. This disclosure would provide a POSA with a strong expectation that an AAVrh.10 vector with the S665N mutation would be viable and could possess enhanced transduction efficiency.
• Additional Grounds: Petitioner asserted further obviousness challenges for claim 8 based on combinations of the '772 Publication with Fabb (a 2002 article on micro-dystrophin), and with Xie and/or Snowdy, to teach the use of a micro-dystrophin gene as the claimed heterologous gene.

4. Key Claim Construction Positions

• For the terms "at least 95% identical" and "at least 97% identical," Petitioner adopted the methodology used by the Patent Owner in co-pending district court litigation. This approach utilizes the Clustal Omega program with default settings to align amino acid sequences and calculate percent identity.

5. Arguments Regarding Discretionary Denial

• Petitioner argued against discretionary denial under Fintiv, stating that the parallel district court litigation is in its early stages. The trial date of November 17, 2025, is approximately 21 months after the petition was filed, ensuring that a Final Written Decision would issue well before trial.
• Petitioner also argued against denial under 35 U.S.C. §325(d). It contended that the key secondary references (Xie, Snowdy, and Fabb) were never before the PTO, and the primary reference (’772 publication), while cited in an Information Disclosure Statement, was never substantively applied or discussed by the examiner during prosecution of the ’274 patent.

6. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1, 3-6, and 8 of Patent 11,680,274 as unpatentable.