Inherency in Obviousness – What is the Correct Standard?

To be awarded a patent, an invention has to be novel and nonobvious and meet a few other requirements.  Obviousness is determined by considering prior art from the standpoint of a hypothetical person of ordinary skill in the art at the time of the invention.  This requires determining whether the ordinarily skilled person would have been able to arrive at the invention with all its features in view of the teachings of the prior art.  It would therefore appear that an invention having a feature that was unknown at the time of the invention (i.e., an inherent feature) could not have been obvious.  “That which may be inherent is not necessarily known. Obviousness cannot be predicated on what is unknown.” In re Spormann, 363 F.2d 444, 448 (C.C.P.A. 1966).  However, reality is more complex.  As stated in M.P.E.P. § 2112, claims may be rejected under both 35 § U.S.C. 102 (novelty) and 103 (obviousness) based on express, implicit, and inherent disclosures of a prior art reference.

Inherent obviousness is a difficult concept and sometimes confused with inherent anticipation.  A claim can be rejected for inherent anticipation when a prior art reference discloses all but one (at least one) limitations – if that limitation necessarily flows from the prior art. M.P.E.P. § 2112 (IV).  The limitation need not, however, have been recognized at the time of the invention, the underlying logic being that prior art that includes all elements of a claim, explicitly or inherently, cannot be practiced without infringing the claim.  And, since a patent claim may not cover that which already exists in the prior art, it is anticipated.  A good illustration of inherent anticipation can be found in Titanium Metals Corp. v. Banner, 778 F.2d 775 (Fed. Cir. 1985).  In Titanium, claims were directed to a titanium alloy that contained defined percentages of Mo and Ni and possessed good corrosion resistance.  The Federal Circuit found the claimed alloy anticipated since the prior art alloy metal percentages were within the metal percent ranges recited in the claim.  The prior art, however, was silent about corrosion resistance, but this fact did not help the patentee.  In the court’s view, as to novelty, it was immaterial what inherent properties the alloys possessed or whether applicants discovered certain inherent properties.

Moving on inherent obviousness, let us first consider In re Napier, 55 F.3d 610, 613 (Fed. Cir. 1995).  In this case, the source of inherency was a known scientific principle.  Napier’s claims were directed to a device that reduced noise from an airplane auxiliary power unit (APU).  An APU is a combustion engine that provides power to the plane while it is on the ground.  The operation of both the Napier device and that disclosed in the prior art was based on the principle of redirection of noise.  The claimed device redirected the APU noise by forcing a stream of cold air through an exhaust pipe (making it less inconvenient for the ground crew servicing the plane).  Cold air being denser than hot engine exhaust, the noise was refracted.  Prior art taught a propulsion jet engine that mixed cold air into its exhaust stream to reduce jet engine noise.  Napier argued – unsuccessfully – that his invention was different because it used ‘‘refraction’’ while the prior art device used ‘‘mixing.’’  Not convinced, the Federal Circuit determined that prior art inherently disclosed redirection of noise and found the claimed device obvious.

How about when the missing and allegedly inherent limitation is not known to the skilled artisan at the time of the invention?  Par Pharmaceutical, Inc. v. TWI Pharmaceuticals, 773 F.3d 1186, (Fed. Cir. 2014) provides a recent example of this scenario.  Par Pharmaceuticals (Par) obtained a license for US patent number 7,101,576 (‘576 patent).  The patent claimed methods of treating certain wasting diseases in humans.  The methods required administering a megestrol acetate nanoparticle formulation.  A further requirement was that there be no substantial difference in C_max according to whether the drug was administered to a patient in the fasting state or the fed state (food effect).  A dependent claim specified an upper limit for the C_max difference.

The claimed formulation was approved for marketing as MEGACE ES.  TWI Pharmaceuticals (TWi) sought to manufacture a generic version of MEGACE ES and was sued by Par for infringement.  At the district court, TWi defended on the basis that claims of the ‘576 patent were invalid as obvious.  Prior art references included MEGACE OS, a micronized form of megesterol acetate (a predecessor to MEGACE ES) as well as a set of patents that taught the so-called NanoCrystal technology for making drug particles.  Prosecution history of these patents indicated that nanoparticle formulations of certain drugs exhibited substantial increases in bioavailability, suggesting that the technology could lead to “decreased fed-fasted variability and more rapid onset of action.”  Another relevant fact was that NanoCrystal technology had been billed as having the potential, among others, to reduce fed-fasted effects and viscosity in poorly soluble drugs.

The district court found that given the known high viscosity and high interpatient variability of MEGACE OS, there was motivation to combine the references for creating a method of treatment using nanoparticles.  The parties agreed that except for the food effect, all other limitations were taught in the prior art.  The prior art did not contain any explicit teaching of the food-effect limitation.  However, TWi and the district court contended that the limitation was an inherent property of the formulation disclosed by the obvious combination of prior art elements.

While agreeing that there was no explicit teaching of the food-effect limitation, the Federal Circuit found the district court’s inherency analysis to be incorrect.  Evidence presented by Twi was found to be directed only at proving that reduction in particle size improved bioavailability and that improvement in bioavailability necessarily resulted in a decrease in any food effect.  However, it was not found to provide clear and convincing evidence that there was “no substantial difference in C_max” between the fed and fasted states, as required by the claims.  The court referred to an earlier opinion that stated, “[t]he mere fact that a certain thing may result from a given set of circumstances is not sufficient. (Citations omitted.) If, however, the disclosure is sufficient to show that the natural result flowing from the operation as taught would result in the performance of the questioned function, it seems to be well settled that the disclosure should be regarded as sufficient.”  Quoting In re Oelrich, 666 F.2d at 581 (C.C.P.A. 1981) (emphases added).  The Federal Circuit remanded the case to the district court to determine if TWi had presented clear and convincing evidence demonstrating that the food effect as claimed was necessarily present in the prior art combination.

Clearly, Par v. Twi, where the inherent limitation was not known, presents a more complex situation than In re Napier, which involved a known scientific principle.  Par v. Twi, however, also raises a question regarding the correct standard for determining inherent obviousness.  In fact, the “?ow naturally from” standard used by the court had been rejected by the CCPA in favor of a “predictability” standard.  See In re Henderson, 348 F.2d 550, 553 (C.C.P.A. 1965).

In Henderson, the rejected claims were directed to a gasoline composition (blend) having branched chain di-lower-alkyl ether added to a particular gasoline mixture.  Surprisingly, although the ether’s sensitivity was higher than that of the base gasoline, it lowered the sensitivity of the base gasoline mixture instead of raising it further, and even more surprisingly, the sensitivity of the blend turned out to be lower than that of either the ether or the gasoline mixture separately.  There was no suggestion in the art that the sensitivity factor of the blend would be lower than the sensitivity factor of either of the individual components.  The Board of Patent Appeals and Interference (“Board”) took the position that the claimed blend was obvious since the ether additive would ‘inherently function’ to produce the decreased sensitivity.  The CCPA reversed, stating, “[t]he mere notion that such discovery would ‘flow naturally’ from what appellant did does not mean that the discovery could be predicted from what the art had done”, asking rhetorically, “[i]ndeed, can it not be said that all functions of a substance are ‘inherent’?”

Thus, Henderson requires asking whether an ordinarily skilled person, at the time of the invention, could have predicted the combination of references to result in the missing limitation (the claimed benefit).  In Par v. Twi, while the Federal Circuit reached the correct result, it used the “naturally results from” test – not the test used in Henderson.  The court based its use of the “naturally results from” test on a CCPA opinion in In re Oelrich (C.C.P.A. 1981), which, in fact, was addressing inherency in anticipation, not obviousness. In re Oelrich, 666 F.2d 578, 580 (C.C.P.A. 1981) (stating “[t]he issue in the former case [In re Oelrich, 579 F.2d 86, 198 USPQ 210 (CCPA 1978)] was obviousness; here it is anticipation.”)

Although the distinction between inherency in obviousness and anticipation is sometimes blurred, the two concepts are quite different and a claim may be inherently anticipated without being inherently obvious.  This could happen if the missing and unknown limitation were to flow naturally from the teachings of the prior art, and yet not be predictedable.  This scenario is well illustrated by Ex parte Novitski, 26 USPQ 2d 1389 (BPAI 1993).  In Novitski the claims at issue were directed to a method for protecting plants from pathogenic nematodes using a nematode-inhibiting strain of Pseudomonas cepacia, the strain colonizing the plant.  The primary reference disclosed a method requiring inoculating a plant with Pseudomonas cepacia type Wisconsin 526, which colonizes the plant and has broad anti-fungal property.  Secondary references only showed methods of testing bacteria for nematode-inhibiting properties.  According to the examiner, the combination of references would have led an ordinarily skilled person to the claimed method with a “reasonable expectation of success.”  Disagreeing, the Board pointed out that the “secondary” references taught screening bacteria for nematode-inhibiting activity and, at most, attributed such activity to several strains of Pseudomonas bacteria which were not Pseudomonas cepacia.  Therefore, the secondary art fell short of establishing that bacterial strains of Pseudomonas cepacia type Wisconsin 526, disclosed by the primary reference, could reasonably be expected to possess nematode-inhibiting activity.  However, the Board sua sponte rejected the claims for anticipation finding that nematode-inhibiting property was inherent to the strain based on disclosure in the Specification that Pseudomonas cepacia type Wisconsin 526 possessed an 18% nematode-inhibition rating.  Applicant argued, unsuccessfully, that the Specification described 40% inhibition rating as being the minimum required for the purposes of the claim.  However, in the Board’s view, Applicant had not established this minimum to be an accurate and appropriate criterion recognized by a person of ordinary skill in the art.

The standard used in Henderson and Novitski, i.e., predictability of arriving at the missing limitation, would seem to be the proper standard for determining inherent obviousness.  Asking if the limitation is “necessarily present” or “naturally flows from” is to address inherent anticipation, not inherent obviousness.  However, M.P.E.P. § 2112 does not articulate the predictability standard.  Should this be taken to mean that there is only one standard for determining inherency in both anticipation and obviousness – the one for determining inherent anticipation?  The answer most likely is “No.”  In Par v. Twi, although the court used the “?ow naturally from” standard, it stressed that inherency was rooted in anticipation and must be carefully circumscribed in the context of obviousness.  Also, more recently the Federal Circuit reversed a district court’s conclusion of obviousness reached on the ground that the invention (a compound) was the “inherent result of an allegedly obvious process”, pointing out that none of the experts “foresaw, or expected” the “long-sought properties and advantages” of the claimed compound.  See Millenium Pharmaceuticals v. Sandoz, Inc. Appeal No. 15-2066, Slip opinion, pages 7 and 17 (Federal Circuit, July 17, 2017).

A major difference between having knowledge of the missing limitation versus not having it is that the knowledge can provide the motivation to combine prior art references.  However, under In re Dillon, 919 F.2d 688 (Fed. Cir. 1990) (en banc), a prima facie case for obviousness does not require the motivation to combine prior art references to match the inventor’s motivation – as long as another plausible motivation is available.  With the references having been thus combined, if the missing (but unknown) limitation naturally flows, the invention apparently is obvious.  Is this a proper analysis of inherency in obviousness?  Or, may one still inquire whether the limitation was within the expectation of the ordinarily skilled person, i.e., apply the predictability standard?  If such inquiry is necessary, then what, if anything, is gained by first applying the “naturally flows” standard?  And, if not, then are we not merely determining inherent anticipation?