PTAB

IPR2018-00739

Moderna Therapeutics Inc v. Arbutus Biopharma Corp

Key Events

Petition

petition

Title: Lipid Formulations for Nucleic Acid Delivery
Brief Description: The ’435 patent describes nucleic acid-lipid particles for delivering therapeutic nucleic acid payloads. The core invention asserted is a specific composition of these particles, defined by claimed molar percentage ranges for three key lipid components: a cationic lipid, a non-cationic lipid, and a conjugated lipid.

Prior Art Relied Upon: The ’196 PCT (WO 2005/007196) and the ’189 publication (Application # US2006/0134189).
Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that the Patent Owner’s own prior art discloses the same three types of lipids (cationic, non-cationic, conjugated) used to form nucleic acid-lipid particles. These references teach broad and preferred component ranges that overlap with or encompass the ranges claimed in the ’435 patent. For example, the ’196 PCT taught a cationic lipid range of 2-60%, which overlaps the ’435 patent’s claimed range of 50-85%. This overlap, Petitioner asserted, establishes a prima facie case of obviousness.
- Motivation to Combine: The motivation was characterized as simple and routine optimization. A person of ordinary skill in the art (POSITA) would have been motivated to vary the proportions of known lipid components within the disclosed ranges to determine the optimal formulation for efficacy, a standard practice in the field.
- Key Aspects: A central argument was that the Patent Owner could not rely on "unexpected results" to overcome obviousness. Petitioner contended that the experimental data within the ’435 patent itself is not commensurate with the scope of the claims and, in some cases, shows that formulations within the claimed ranges perform worse than admitted prior art formulations.

Prior Art Relied Upon: The ’196 PCT and ’189 publication, in view of Lin (a 2003 Biophysical Journal article) and/or Ahmad (a 2005 Journal of Gene Medicine article).
Core Argument for this Ground:
- Prior Art Mapping: This ground supplements Ground 1. To the extent the base references are deemed insufficient to suggest the higher cationic lipid concentrations, Petitioner argued that Lin and Ahmad explicitly teach the benefits of using cationic lipid proportions at or above 50 mol%. Both references studied how transfection efficiency changes with the molar fraction of cationic lipid. Ahmad, for instance, showed monotonically increasing efficiency for the DOTAP lipid up to 90 mol%, while other lipids showed peak efficiency around 50-55 mol%.
- Motivation to Combine: A POSITA, seeking to optimize the lipid nanoparticle formulations disclosed in the ’196 PCT and ’189 publication, would have been motivated to consult teachings like Lin and Ahmad to understand the effect of component ratios on transfection efficiency. These references would have directly suggested exploring cationic lipid concentrations in the 50-85% range to improve performance.
- Expectation of Success: Because Lin and Ahmad demonstrated that increasing cationic lipid concentrations above 50% could, for certain lipids, enhance transfection efficiency, a POSITA would have had a reasonable expectation of success in applying this principle to the known formulations.

Prior Art Relied Upon: The ’554 publication (Application # US2006/0240554).
Core Argument for this Ground:
- Prior Art Mapping: Petitioner asserted that the ’554 publication, another of the Patent Owner's prior disclosures, anticipates every limitation of claim 1. The reference discloses the same three lipid components and discloses ranges that overlap the challenged claims (e.g., cationic lipid from 2-60%). Critically, Petitioner pointed to a specific formulation example, L054, in the ’554 publication which contains 50% cationic lipid, 48% non-cationic lipid, and 2% conjugated lipid. Petitioner argued this single example falls squarely within the ranges of claim 1 of the ’435 patent (50-85% cationic, 13-49.5% non-cationic, 0.5-2% conjugated), thus anticipating the claim.
- Motivation to Combine (for Obviousness): In the alternative, Petitioner argued for obviousness based on the overlapping ranges disclosed in the ’554 publication, establishing a prima facie case. The motivation would be the routine optimization of lipid ratios for a known purpose. A POSITA would have been motivated to increase the cationic lipid percentage toward the upper end of the disclosed range (e.g., 60%) to increase the efficiency of a pH-dependent phase shift described in the reference, leading directly to the claimed composition.

"Nucleic acid-lipid particle": Petitioner proposed this term be construed as "a composition of lipids and a nucleic acid for delivering a nucleic acid to a target site of interest," based on the patent's own description.
"Cationic Lipid": Petitioner proposed this term means "any of a number of lipid species that carry a net positive charge at a selected pH, such as physiological pH (e.g., pH of about 7.0)."
Importance: These proposed constructions, drawn directly from the specification, are broad. Petitioner leveraged this breadth to argue that the limited experimental data in the patent, which tested only one specific cationic lipid (DLinDMA), was insufficient to support the patentability of claims covering hundreds of potential lipid species.

Petitioner’s primary technical contention was that the ’435 patent’s own intrinsic evidence fails to support its claims of "unexpected results." Petitioner analyzed the patent’s experimental data (Examples 2-4) to argue that certain formulations falling within the claimed ranges performed no better, or even worse, than prior art formulations. For example, in in-vivo testing, Group 14 (a claimed formulation) appeared to perform worse than Group 7 (an admitted prior art formulation). This alleged lack of consistent superiority across the entire claimed range was used to attack the non-obviousness of the invention.

Petitioner requested the institution of an inter partes review and the cancellation of claims 1-20 of Patent 9,364,435 as unpatentable.