Obvious to Try

The obvious-to-try doctrine holds that choosing among a finite, known set of options is not inventive: DOCTRINE: an invention is obvious if, at the time of the invention, a person of ordinary skill in the art (POSITA) would have: (a) recognized a finite number of identified, predictable solutions to the problem; AND (b) had a reasonable expectation of success from trying each; if these conditions are met, systematically trying the options and discovering the one that works is 'obvious to try' and thus not patentable; KSR INTERNATIONAL v. TELEFLEX (S.Ct. 2007): the Supreme Court confirmed the obvious-to-try rationale is a valid basis for finding obviousness; the Court explicitly rejected the Federal Circuit's strict 'teaching, suggestion, or motivation' (TSM) test as the sole basis for finding a motivation to combine; KSR: 'A person of ordinary skill is also a person of ordinary creativity, not an automaton'; WHEN OBVIOUS-TO-TRY APPLIES (KSR): 'if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious unless its actual application is beyond his or her skill'; 'obvious to try' specifically applies when: there is a design need or market pressure; there are a finite number of identified, predictable solutions; a person of ordinary skill has reason to pursue those solutions; CRITICAL LIMITATION: obvious-to-try does NOT apply where: the results are unpredictable (especially in pharmaceutical/biological arts); the field is highly complex with unpredictable outcomes; the choice was from a very large number of possibilities; the prior art teaches away from the chosen option.

USPTO examiners regularly apply obvious-to-try as a rejection rationale in pharmaceutical, chemical, and biotechnology applications: EXAMINER'S BURDEN: the examiner must show: (1) that there was a finite, limited number of identified predictable solutions; (2) that a POSITA had reason to try each of those solutions; (3) that a POSITA would have had a reasonable expectation of success in achieving the claimed result; COMMON OBVIOUS-TO-TRY SCENARIOS: PHARMACEUTICAL FORMULATIONS: a prior art drug compound with a known pharmacological activity; the examiner argues it was obvious to try salt forms, prodrugs, polymorphs, or dosage adjustments; CHEMICAL COMPOUNDS: selecting a substituent from a closed Markush group (e.g., substituting H, CH3, or OH at a particular position); RANGES AND OPTIMIZATION: optimizing a variable within a known range when the result is predictably better; selecting a temperature, pressure, or concentration from a disclosed range; OBVIOUSNESS 'BY OBVIOUS-TO-TRY': examiner cites prior art disclosing the genus and argues the specific claimed compound was obvious to try; IN RE O'FARRELL (Fed. Cir. 1988): established that when the prior art teaches a clear course of conduct, 'obvious to try' is a valid basis for rejection even without explicit teaching of the specific combination; USPTO 2019 REVISED GUIDANCE: obvious-to-try is one of several established rationale for motivation to combine; listed in the MPEP as a proper basis for § 103 rejection; RESPONSE STRATEGY: applicant should argue: (a) the results were unpredictable; (b) the number of options was enormous, not finite; (c) prior art taught away from the specific choice; (d) secondary considerations show actual non-obviousness.

The obvious-to-try doctrine has important limitations that prevent it from swallowing all pharmaceutical and chemical innovation: UNPREDICTABILITY LIMIT: the most important limitation; obvious-to-try does NOT apply when the results of trying the options are not predictable; in pharmaceutical and biological arts, results are often unpredictable — a structural change may improve or destroy efficacy, cause unforeseen side effects, or have no effect; EISAI CO. v. DR. REDDY'S LABS (Fed. Cir. 2008): a compound is not obvious-to-try when there is no 'lead compound' that a POSITA would have selected to modify AND the prior art does not teach which modifications to make and why; LEAD COMPOUND DOCTRINE IN PHARMACEUTICALS: for pharmaceutical obviousness, courts ask: (a) is there a 'lead compound' in the prior art that a POSITA would have selected for modification?; (b) would a POSITA have had reason to modify the lead with a reasonable expectation of success?; both prongs must be satisfied; LARGE NUMBER OF OPTIONS: if the number of combinations or options is enormous (many thousands or millions), the set is NOT 'finite and identifiable' for obvious-to-try purposes; a Markush group of 10^15 compounds is not finite in the relevant sense; TEACHING AWAY: if prior art teaches away from the specific choice (warns against it, shows poor results), the obvious-to-try doctrine typically does not apply; UNEXPECTED RESULTS: even if a result was obvious-to-try, unexpected results of the specific choice can rebut the obvious-to-try rationale and support patentability; KSR ITSELF: the Supreme Court's flexible approach in KSR did not eliminate unpredictability as a defense — it rejected rigid formalism but recognized that predictability is central to obvious-to-try.

In software and technology arts, obvious-to-try typically arises in the context of combining known techniques or applying known techniques in new contexts: SOFTWARE AND TECHNOLOGY OBVIOUS-TO-TRY: KSR explicitly addressed design engineering: when there is a design need and a finite number of known solutions, combining them is obvious; EXAMPLE SCENARIOS: combining two known algorithms to solve a problem (e.g., combining known encryption with known compression); using a known UI pattern from mobile apps in a new context; adapting a known network protocol to a new data type; OBVIOUS-TO-TRY IN SOFTWARE IS DIFFERENT FROM PHARMA: software results are often more predictable than chemical or biological outcomes; if a programmer would predictably expect that combining two known techniques would work as claimed, obvious-to-try is more likely; EVIDENCE OF UNPREDICTABILITY: applicants in software can still argue: (a) the combination required non-obvious technical insight (specific interaction was not predictable); (b) the result achieved (performance improvement, new functionality) was unexpected; (c) prior art taught away from the combination; (d) skilled artisan would not have expected the combination to work; KSR 'OBVIOUS-TO-TRY' IN MACHINE LEARNING: as machine learning techniques became established prior art, examiners increasingly argue it was 'obvious to apply ML' to a new domain; RESPONSE: the application of ML was not obvious because: (a) it was not clear ML would work in this domain; (b) specific feature engineering or architecture choices were non-obvious; (c) training data or label strategy was innovative; ALICE/§ 101 INTERACTION: in software, § 101 eligibility often overlaps with the § 103 analysis; if the claimed steps are all conventional, both § 101 and obvious-to-try § 103 arguments are available to examiners.

Overcoming obvious-to-try rejections requires attacking either the 'finite options' or 'reasonable expectation of success' prongs: STRATEGY 1 — ATTACK 'FINITE AND PREDICTABLE' OPTIONS: argue the prior art discloses an enormous range of options, not a finite, identified set; show that the relevant parameter space was very large and there was no reason to select the specific option claimed; distinguish between 'could try' and 'would try' — a POSITA could theoretically try millions of options but had no reason to try this specific one; STRATEGY 2 — ATTACK 'REASONABLE EXPECTATION OF SUCCESS': argue that the art is highly unpredictable in the relevant respect; present evidence that skilled artisans in the field would not have expected the claimed result (scientific literature, expert declarations); if prior art shows mixed results for similar approaches, argue unpredictability; STRATEGY 3 — TEACHING AWAY: show prior art explicitly or implicitly taught away from the combination; prior art showing poor results, warnings, or negative guidance regarding the claimed approach undermines obvious-to-try; STRATEGY 4 — SECONDARY CONSIDERATIONS: commercial success (with nexus), long-felt unmet need (others tried and failed with obvious options), failure of others, unexpected results, and industry praise can rebut an obvious-to-try finding; UNEXPECTED RESULTS: perhaps the most effective rebuttal; file a § 1.132 declaration by the inventor or expert showing that the claimed approach produced results that were unexpected compared to the closest prior art; quantitative data comparing the claimed approach to obvious alternatives strengthens the declaration; STRATEGY 5 — NARROW CLAIMS: if broad independent claims are obviously-to-try, narrow to specific embodiments with unexpected results that support patentability; sacrifice breadth for allowability.