Unexpected Results

Unexpected results are a type of secondary consideration (also called objective indicia of non-obviousness) that can rebut a prima facie case of obviousness under 35 U.S.C. § 103: LEGAL BASIS: Graham v. John Deere Co. (S.Ct. 1966) established the four-factor obviousness analysis and held that secondary considerations — including unexpected results — must be considered before a final obviousness determination; KSR International Co. v. Teleflex Inc. (S.Ct. 2007): reaffirmed that secondary considerations must be considered but applied a more flexible combination analysis; WHAT QUALIFIES AS UNEXPECTED: the results must be SURPRISING to a person of ordinary skill in the art (POSA) — not just better, but better in a way that the POSA would not have expected; UNEXPECTED IN DEGREE: performance dramatically and unexpectedly better than the closest prior art — e.g., compound X in the claimed formulation works 10× better than the prior art would predict; courts generally require the difference to be more than a small, incremental improvement; UNEXPECTED IN KIND: the claimed invention produces a result of a different type from what was expected — e.g., a claimed compound exhibits antifungal activity when prior art compounds of similar structure had only antibacterial activity; different-in-kind results are stronger evidence of non-obviousness than different-in-degree results.

Unexpected results must be supported by concrete, comparative evidence: REQUIRED ELEMENTS: (1) COMPARISON TO CLOSEST PRIOR ART: the comparison must be to the CLOSEST prior art (most structurally similar compound, most similar composition, nearest combination), not just any prior art compound; a comparison to a distant or weakly similar reference provides little support; (2) SIDE-BY-SIDE COMPARISON: ideally, the applicant shows a direct head-to-head test of the claimed invention vs. the closest prior art under identical conditions; (3) STATISTICAL SIGNIFICANCE: the data must show a statistically significant difference; anecdotal or single-data-point evidence is weak; (4) INTERPRETATION BY SKILLED ARTISAN: a declaration explaining why a skilled artisan would find the result surprising is standard practice; FORM OF EVIDENCE: declarations under 37 C.F.R. § 1.132 are the primary vehicle for establishing unexpected results in prosecution; inventors, technical experts, or other qualified persons can submit declarations; data tables, graphs, and experimental protocols should be attached; TIMING: unexpected results evidence can be submitted after an obviousness rejection, but must be submitted before a final rejection becomes final/abandoned; evidence of record at the time of prosecution is available in litigation as well; QUALITY OF COMPARISON: all variables other than the claimed feature should be held constant; changing multiple variables makes it impossible to attribute the result to the specific claimed element.

Unexpected results are most frequently litigated in the chemical and pharmaceutical arts: CHEMICAL PATENTS — STRUCTURAL ANALOGS: when a claimed compound is structurally similar to prior art compounds, an examiner often rejects for obviousness by arguing that analogs are routinely synthesized; unexpected results can overcome this by showing the claimed compound has properties that prior art analogs of the same class do not have; the closer the structural similarity, the more dramatic the unexpected results must be; LEAD COMPOUND ANALYSIS: Federal Circuit often applies a 'lead compound' analysis for pharmaceutical obviousness — was there a lead compound to modify, and would a skilled artisan have been motivated to make the modification? If the modification was unexpected, it weighs against motivation or predictability; SYNERGISTIC EFFECTS: in combination patent cases, if compounds A + B together produce results unexpectedly better than the sum of A and B alone, synergy is a form of unexpected result that strongly supports non-obviousness; however, synergy must be proven, not merely asserted; PHARMACEUTICAL FORMULATION: if a formulation (e.g., specific excipient, co-crystal, polymorph) produces unexpected bioavailability, solubility, or stability improvements, this can be an unexpected result distinguishing the formulation from obvious alternatives; METABOLITE CLAIMS: discovering that a metabolite of a known drug has its own potent activity is a type of unexpected result, though it is also subject to inherency challenges (the metabolite was always produced — was it unexpectedly beneficial?).

Unexpected results are important but not automatically decisive — courts evaluate them carefully: LIMITATIONS: (1) NEXUS REQUIREMENT: the unexpected results must be CAUSED BY the features that distinguish the claimed invention from the prior art; if the unexpected result is attributable to a non-claimed feature or a feature common with the prior art, there is no nexus and the evidence is discounted; (2) BREADTH OF CLAIM vs. DATA: the data must support the FULL SCOPE of the claim; if claims cover a range or genus, testing only one species or one point in the range may not support the entire claim; Genetics Institute v. Novartis Vaccines (Fed. Cir. 2011): representative testing can establish unexpected results for a genus if the evidence reasonably predicts results across the genus; (3) WEAK MAGNITUDE: 'better' is not enough — 'unexpectedly better' is required; a 15% improvement in a field where 10–20% variation is expected is probably not unexpected; (4) SELF-SERVING DECLARATIONS: examiner and courts may discount self-serving declarations from inventors without independent supporting data; third-party testing is more persuasive; (5) PRIMA FACIE STRENGTH: the stronger the examiner's prima facie case (e.g., prior art teaches the exact combination), the more persuasive the unexpected results must be; (6) BALANCING WITH OTHER SECONDARY CONSIDERATIONS: unexpected results are ONE factor — courts also weigh long-felt need, failure of others, commercial success, copying, and teaching away; a weak unexpected results showing may still be sufficient if supported by other secondary considerations.

Unexpected results arguments play out somewhat differently in prosecution vs. litigation: PROSECUTION: applicant submits declarations with comparative data; the examiner evaluates whether the comparison is to the closest prior art and whether results are truly unexpected; if persuasive, the rejection is overcome; the record of arguments and data submitted during prosecution becomes part of the file history; LITIGATION (IPR AND DISTRICT COURT): patent owner asserts unexpected results as evidence of non-obviousness; clear and convincing evidence standard for invalidity in district court; preponderance standard in IPR; the PTAB applies strict nexus requirements — the patent owner must show the unexpected results are coextensive with the claim scope (not just a subset of the claims); COMPARING SCOPE TO EVIDENCE: if the claim is broad (e.g., genus of 1,000 compounds) and testing covers only 5 compounds, the PTAB may find the evidence insufficient to establish unexpected results across the full claim scope; in district court, a jury may be more receptive to expert testimony interpreting data; TEMPORAL CONSIDERATIONS: in prosecution, unexpected results data can be submitted in response to a rejection; in litigation, the patent owner presents evidence at trial — but if the evidence was not part of the prosecution record, it can be submitted at trial; WEIGHT: courts consistently hold that unexpected results, even when established, do not automatically overcome a strong prima facie case; they are weighed in the totality of secondary considerations evidence.