Pfizer v. Sanofi: Applying the Results-Effective Variable Doctrine in Obviousness Analysis

by Dennis Crouch

The Federal Circuit has affirmed the PTAB’s finding that Pfizer’s pneumococcal vaccine patent is obvious, but has vacated and remanded the Board’s denial of Pfizer’s motion to amend certain claims. Pfizer Inc. v. Sanofi Pasteur Inc., No. 19-1871 (Fed. Cir. Mar. 5, 2024); U.S. Patent No. 9,492,559. Pfizer v. Sanofi Opinion.

The ‘559 patent claims immunogenic compositions comprising glycoconjugates of various Streptococcus pneumoniae serotypes for use in pneumococcal vaccines. Pneumococcal diseases such as pneumonia, bacteremia, and meningitis are a significant public health concern, and effective vaccines against the most common S. pneumoniae serotypes are crucial.

Sanofi Pasteur and Merck Sharp & Dohme (the “petitioners”) challenged the validity of the ‘559 patent claims in five inter partes review (IPR) proceedings, arguing they were obvious over prior art references disclosing multivalent pneumococcal vaccines. The PTAB agreed, finding all challenged claims unpatentable.  On appeal, Pfizer challenged the Board’s obviousness determinations and its denial of Pfizer’s contingent motion to amend.

Results-Effective Variable Doctrine: In its decision, the PTAB applied the “result-effective variable doctrine,” that is generally applied in claims of ranges or particular parameters.  Under the doctrine, prior art that that discloses the “general conditions of a claim” will create a presumption of obviousness if the particular workable ranges are identifiable through routine experimentation. See In re Aller, 220 F.2d 454 (CCPA 1955).  This doctrine is often applied where a claim recites a numerical range for a parameter, and the prior art discloses a broader or overlapping range for that same parameter.  Here, the court wrote that “an overlap between a claimed range and a prior art range creates a presumption of obviousness that can be rebutted with evidence that the given parameter was not recognized as result-effective. See Genentech, Inc. v. Hospira, Inc., 946 F.3d 1333 (Fed. Cir. 2020).  Evidence that a particular claimed range is not “result-effective” would show that the parameter affects the properties or function of the claimed invention in an unpredictable manner.

In the present case, a key issue was whether the claimed molecular weight range of 1,000 to 12,500 kDa for the serotype 22F glycoconjugate would have been obvious, where the prior art did not expressly disclose that range. The Federal Circuit agreed with the Board that the prior art rendered this limitation obvious under the result-effective variable doctrine.  Although the exact claimed range was not disclosed, the prior art did show molecular weight ranges for 14 other S. pneumoniae serotype glycoconjugates that overlapped with the claimed range. The court also pointed to teachings in the prior art that glycoconjugate molecular weight affects conjugate stability and immunogenicity in a predictable manner.  As the court explained:

[The] evidence therefore supports the Board’s conclusion that “conjugate size is a result[-]effective variable associated with improved stability of conjugates and good immune response, limited only by filter size, thereby rendering ‘optimization within the grasp of one of ordinary skill in the art.’” (quoting Applied Materials).

Pfizer argued that the doctrine should not apply where there is no express range disclosed in the prior art for that specific parameter (i.e., the molecular weight of serotype 22F). But the Federal Circuit disagreed, explaining that the doctrine does not require an express disclosure of the parameter range in the prior art, as long as the parameter is recognized as result-effective.

Applying the result-effective variable doctrine here significantly strengthened the obviousness case, as it avoided the need for the prior art to disclose the exact claimed range. The doctrine allowed the Board and court to find this claim limitation obvious based on the general disclosure of overlapping ranges for related serotypes and the recognition that the parameter could be optimized predictably.

In its analysis, the Federal Circuit noted that “the fact that the art of pneumococcal glycoconjugate vaccines is unpredictable does not affect our analysis” because this particular parameter (molecular weight) was recognized in the expert testimony as being stable across similar serotypes.

Pfizer also argued there was insufficient evidence of a reasonable expectation of success in achieving the claimed immunogenicity for the glycoconjugates of additional serotypes recited in dependent claims. Again, the Federal Circuit disagreed, emphasizing that “obviousness cannot be avoided simply by a showing of some degree of unpredictability in the art so long as there was a reasonable probability of success.” Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348 (Fed. Cir. 2007).  The following comes from oral arguments between Judge Lourie and of the Judges and the patent challenger’s attorney Siegmund Gutman.

Judge Lourie: What about the unpredictability in the field that your opposing counsel talked about?

Gutman: Pfizer made the argument, that precise argument that is in their opening brief, in the reply brief, and that my good friend made here earlier, that the immunogenic composition development field, multivalent PCV compositions is unpredictable. That argument was made by the Board. It was considered by the Board. It was found to be unpersuasive by the Board.

In the end, the court reaffirmed that “an expectation of success need only be reasonable, not absolute.” Quoting Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348 (Fed. Cir. 2007)

Turning to Pfizer’s motion to amend, the Federal Circuit affirmed the Board’s denial as to proposed claim 46, but vacated and remanded as to proposed claims 48 and 49. Those claims required the immunogenic composition exhibit a specific increase in IgG antibody levels across all serotypes. When a patent owner moves to amend claims in an IPR, the Board evaluates the proposed amended claims to determine if they are patentable over the prior art of record. The patent owner bears the burden of showing that the proposed amended claims are patentable.  During that process, the Board must assess the patentability of each proposed claim individually, determining whether the patent owner has carried its burden to show that each claim is patentable. The Board should provide a reasoned basis for either allowing or rejecting each proposed claim.

The court found the Board failed to adequately explain why the additional serotypes recited in claims 48 and 49 would have been expected to exhibit the claimed 2x increase in antibodies, stating “we cannot say, with any modicum of certainty” that the Board’s conclusion was supported by substantial evidence.

The key principle is that the Board, when denying a motion to amend, must set forth a reasoned basis for why each proposed claim is unpatentable. Merely finding one claim obvious (like claim 46) does not automatically render all dependent claims obvious without a claim-specific analysis — especially as here where the other claims were directed to a different serotype.

Direct Review APA Claim: Finally, the court rejected Pfizer’s argument that the PTO’s Director Review procedure implemented post-Arthrex violated the Administrative Procedure Act (APA). Even assuming the PTO’s guidance was subject to notice-and-comment rulemaking, the court held any APA violation was harmless because Pfizer failed to show it was prejudiced.

While Pfizer’s ‘559 patent remains invalid, the case will now return to the PTAB for further consideration of whether proposed claims 48 and 49 are independently patentable.

= = =

Opinion by Judge Lourie, joined by Judge Bryson and Stark.

For the Appellant, Pfizer Inc.: John Scheibeler from White & Case argued the appeal.  He was joined on the brief by a host of White & Case lawyers, including Dimitrios Drivas, Amit Thakore, Elizabeth Chang, Catalin Zonte, and Henry Huang. On the other side, Siegmund Gutman from Proskauer Rose argued on behalf of petitioners.  He was joined on the brief by his partner John E. Roberts.  For the USPTO, Mary L. Kelly argued and was joined with others from the USPTO Solicitors office, including Peter Ayers, Daniel Kazhdan, and Farheena Yasmeen Rasheed, along with Scott McIntosh from the USDOJ civil division.

13 thoughts on “Pfizer v. Sanofi: Applying the Results-Effective Variable Doctrine in Obviousness Analysis

  1. 5

    “See In re Aller, 220 F.2d 454 (CCPA 1955). ”

    In my experience, Aller is one of the most highly abused cases cited by Examiners. The most recent experience was with an Examiner who cited a numeric value disclosed in the prior art that was well below the range claimed by the Applicant. No analysis explaining how Aller was being applied, and no evidence provided that routine experimentation would have resulted in the claimed range.

  2. 4

    OT [but an important client-counseling topic infrequently discussed on patent blogs], Gene’s patent blog has an article asserting that millions of dollars are being wasted by automobile companies failing to adequately conduct patent maintenance fee non-payment reviews of their patent portfolios for patents no longer of strategic interest. [It is not just that industry.]

    1. 4.1

      I thought that was interesting as well.

      That being said, HOW the valuations are performed, and how well any identified metrics – especially any forward looking ones – are the Devil in the Details.

    1. 3.1

      Bravo – note that this occurs after peer reviews (for the most part).

      Academia has been rotted to the core for quite some time now, and excellence emerges despite rather than because of the environment there.

    2. 3.3

      Thanks Night.

      Academia needs more Biks. Plenty more.

      Calling 60 minutes. Calling 60 minutes.

  3. 2

    In the end, the court reaffirmed that “an expectation of success need only be reasonable, not absolute.” Quoting Pfizer, Inc. v. Apotex, Inc., 480 F.3d 1348 (Fed. Cir. 2007)
    To repeat a question I asked several articles ago … expectation of success as to what? Who decides whether this unknown something is successful? Is there is standard?

    1. 2.1

      The only standard is our “leans-to-invalidation,” better-than-voodoo crystal ball.

      So saith the CAFC.

  4. 1

    “The patent owner bears the burden of showing that the proposed amended claims are patentable.”

    No. That hasn’t been true since the en banc Aqua Products decision in 2017. PTO then issued regulations expressly putting the burden of persuasion on patentability on petitioners. 37 CFR 42.121(d)(2).

    1. 1.2

      Important aspect to this (before and after).

      No one in a post grant review is actually an examiner.

      Any amended claims (unless merely amending an existing claim to be, say, independent) is being turned around and let loose without an actual examination.

Comments are closed.