The results-effective variable doctrine is a caselaw principle where prior art disclosing the “general conditions of a claim” invokes a presumption of obviousness if the particular workable ranges are identifiable through routine experimentation. The presumption can be rebutted with evidence that the given parameter was not recognized as being result-effective. Sanofi Pasteur, Inc. and SK Chemicals Co initiated an IPR (inter partes review) to invalidate U.S. Patent No. 9,492,559 (the “‘559 Patent”) owned by Pfizer based on the routine optimization theory. The ‘559 Patent is directed to compositions comprising a Streptococcus pneumoniae serotype 22F glycoconjugate with a molecular weight between 1,000 kDa and 12,500 kDa. The IPR petitioners cited two references mentioning the S. pneumoniae 22F serotype, but neither reference disclosed any molecular weight of the claimed glycoconjugate. Notwithstanding the lack of a disclosure of a general range, the Board applied the “results-effective variable doctrine” and found that it would have been obvious to arrive at the claimed invention through routine optimization.

Pfizer appealed the Board’s finding on the grounds that the results-effective variable doctrine cannot apply when there is no actual or implied overlap between a claimed range and a range disclosed in the cited art. The issue before the Federal Circuit was whether the claimed molecular weight range of 1,000-12,500 kDa for the 22F serotype would have been obvious when the cited art did not disclose any range, let alone a range that overlapped with the claimed range, for the reported glycoconjugate. The Court affirmed the Board’s finding of obviousness.

The Court noted that although an overlap of ranges between the claims and the art creates a rebuttable presumption of obviousness, the lack of an overlap does not preclude application of the results-effective variable doctrine. When a routine optimization analysis is applied, the issue “requires whether a person of ordinary skill in the art would have been motivated, with a reasonable expectation of success, to bridge any gaps in the prior art to arrive at a claimed invention.” The fact that a gap is not disclosed in the cited art does not foreclose the routine optimization analysis. Instead, the question is whether the person of ordinary skill in the art would have recognized that the parameter at issue was a result-effective variable. Here, the court agreed with the Board’s conclusion that “substantial evidence” supported the Board’s finding of obviousness, particularly since the cited art acknowledged that thirteen “other” related glycoconjugates had a molecular weight within the claimed 1,000-12,5000 kDa range and no serotype glycoconjugates were reported as having molecular weights outside the claimed range.  

Examiners are quite liberal in applying the routine optimization analysis set forth in In re Aller (220 F.2d 434 (CCPA 1955)), and this CAFC decision will further bolster their application of the routine optimization analysis. Claims in any technology reciting specific ranges can nonetheless be considered obvious if one of ordinary skill in the art would reasonably believe that the parameter at issue is a results-effective variable and that the claimed parameter values can be achieved through routine optimization. Notably, identifying lack of a general range in a cited document may be insufficient to overcome a routine optimization rejection for a range limitation. Therefore, it will be important to include alternative elements to distinguish between the claims and the cited art in case that the range limitation alone is found insufficient. Also, a patent application or a response to an obviousness rejection that invokes a routine optimization analysis needs to include evidence or rationale that the parameter was not known to be result-effective in order to rebut the presumption of obviousness under the results-effective variable doctrine.