PTAB

IPR2018-01229

Merck Sharp & Dohme Corp. v. GlaxoSmithKline Biologicals SA

Log In

1. Case Identification

2. Patent Overview

  • Title: Process for Conjugating Bacterial Saccharides for Vaccines
  • Brief Description: The ’645 patent is directed to a process for making vaccine components by conjugating a specific bacterial saccharide (S. pneumoniae capsular saccharide 23F) to a carrier protein. The claimed invention involves using a specific low concentration range of periodate as an oxidizing agent, which purportedly reduces the "sizing effect"—the fragmentation and size reduction of the saccharide that can occur during the conjugation process.

3. Grounds for Unpatentability

Ground 1: Claims 1-11 are anticipated by WO’376.

  • Prior Art Relied Upon: WO’376 (PCT Application # WO 2004/043376A2).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that WO’376, which was not considered during prosecution, discloses every element of the challenged claims. Specifically, Example 4 of WO’376 describes a method for conjugating bacterial saccharides, including the claimed 23F serotype. It teaches reacting the saccharide with 0.31 molar equivalents (MEq) of periodate, a value falling squarely within the ’645 patent's claimed range of 0.001-0.7 MEq. The reaction occurs in a 100 mM phosphate buffer, which meets the claimed buffer concentration of 1-100 mM. WO’376 further discloses all subsequent steps, including mixing the activated saccharide with a carrier protein and reacting the mixture with a reducing agent (sodium cyanoborohydride) to form the final conjugate.
    • Key Aspects: Petitioner contended that the preamble phrase "reducing the sizing effect" is an inherent and expected result of practicing the steps disclosed in WO’376, particularly the use of a low periodate concentration. As this is an inherent property of a disclosed prior art process, Petitioner argued it cannot impart patentability.

Ground 2: Claims 1-11 are obvious over WO’376 in view of Frasch and Lees.

  • Prior Art Relied Upon: WO’376 (PCT Application # WO 2004/043376A2), Frasch (a 2009 article in Vaccine), and Lees (a 2008 book chapter on conjugation chemistry).

  • Core Argument for this Ground:

    • Prior Art Mapping: Petitioner asserted that even if WO’376 does not anticipate every limitation, the claims would have been obvious in light of its teachings combined with the state of the art as described by Frasch and Lees. WO’376 provides the foundational process for 23F conjugation using periodate. Frasch and Lees explain the known consequences of this chemistry, teaching that periodate activation can fragment the polysaccharide ("sizing effect") and that this degradation is dependent on the periodate concentration. Higher concentrations cause more fragmentation.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSA) would combine these teachings. A POSA starting with the WO’376 process would consult general knowledge references like Frasch and Lees to optimize the reaction. These references explicitly teach that preserving saccharide structure and important epitopes is critical for immunogenicity. They would therefore motivate a POSA to use mild conditions, such as the low periodate concentration taught in WO'376, to achieve the known and desired goal of "reducing the sizing effect" and preserving the final conjugate's effectiveness.
    • Expectation of Success: A POSA would have had a reasonable expectation of success because the principle that lower reactant concentrations lead to less degradation was a fundamental and well-understood concept in chemistry. Applying this principle to the WO'376 process to reduce saccharide fragmentation was a routine and predictable optimization.

  • Additional Grounds: Petitioner asserted additional obviousness challenges against dependent claims, building on the WO'376, Frasch, and Lees combination. These grounds argued that selecting specific molecular weight ranges (claims 4-5), carrier proteins like CRM197 (claim 6), or additional unconjugated antigens (claim 10) would have been obvious modifications based on further teachings from GSK 2009 PCT (WO 2009/000825A2) and Prevnar (a 2009 Physician's Desk Reference).

4. Key Claim Construction Positions

  • "reducing the sizing effect": Petitioner argued this preamble term is not a limitation on the claims but merely a statement of the intended purpose or natural result of performing the claimed steps. Alternatively, if the Board finds the term limiting, Petitioner argued it should be construed by its plain meaning as "decreasing the reduction in the size of the bacterial saccharide," an outcome inherent to using the low periodate concentrations of the claimed process.
  • "molar equivalents": Petitioner argued this term should be construed as "the ratio of moles of periodate to the moles of saccharide repeating unit." This construction is based on calculations derived from examples in the ’645 patent itself, which show that the Patent Owner used the molecular weight of the 23F repeating unit (RU) to arrive at its stated MEq values.
  • "molecular weight": Petitioner contended that the molecular weight (MW) ranges recited in dependent claims 4 and 5 are non-limiting statements of intended results. Alternatively, if found to be limiting, they should be construed as the weight-average MW of the activated saccharide measured prior to conjugation, as defined in the patent's specification.

5. Key Technical Contentions (Beyond Claim Construction)

  • Rebuttal of Patent Owner’s "Unexpected Results" Evidence: A central part of Petitioner's case was a technical attack on the evidence of "unexpected results" that the Patent Owner used to overcome an obviousness rejection during prosecution. Petitioner argued the data in the ’645 patent fails to show that the claimed periodate range is critical or produces unexpected results for several reasons:
    • The experiments in the patent’s Example 1 do not cover the entire claimed range of 0.001-0.7 MEq, providing data for only a narrow slice (0.1-0.5 MEq for 23F).
    • The experiments were not properly controlled, as reaction conditions (e.g., buffer type and concentration) were varied between data points, making it impossible to isolate the effect of the periodate concentration.
    • The data merely shows a predictable, linear trend—that lower periodate concentration leads to less size reduction—which is an expected result, not a surprising discovery or a difference in kind.

6. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-11 of Patent 8,753,645 as unpatentable under 35 U.S.C. §§ 102 and 103.