Trying to Understand What's Not Obvious about What's "Obvious to Try"

    By Kevin E. Noonan --

The Supreme Court's recent KSR Int'l Co. v. Teleflex Inc. decision is stunning in the amount of dicta it contains (and confusing dicta at that).  The Court succumbed to imprecise language (albeit not quite gobbledygook) even when addressing the merits of the case, particularly its reasoning for reversing the Federal Circuit's opinion.  Nowhere is this more evident than in its foray into the question of when something that is "obvious to try" is also obvious under 35 U.S.C. § 103.

This portion of the Court's opinion consists of a mere 124 words, but has raised the temperature of commentators and less measured members of the patent bar, who opine that it somehow lowers the bar for courts (and more fearfully, the Patent and Trademark Office) to find inventions obvious:

The same constricted analysis led the Court of Appeals to conclude, in error, that a patent claim cannot be proved obvious merely by showing that the combination of elements was "obvious to try."  Id., at 289 (internal quotation marks omitted).  When there is a design need or market pressure to solve a problem and there are a finite number of identified, predictable solutions, a person of ordinary skill has good reason to pursue the known options within his or her technical grasp.  If this leads to the anticipated success, it is likely the product not of innovation but of ordinary skill and common sense.  In that instance the fact that a combination was obvious to try might show that it was obvious under § 103.

The take-home message for many are the words "error" in conjunction with "that a patent claim cannot be proved obvious merely by showing that the combination of elements was 'obvious to try.'"  Having created the impression that something that is "merely . . . obvious to try" can be obvious, the Court then contradicts its own statement by qualifying it:  something that is "[m]erely . . . obvious to try" is also obvious only when there is a design need or market pressure (or, in the words of the TSM test, "motivation") and "a finite number of identified, predictable solutions" that "leads to the anticipated success."

Part of the difficulty in explicating the KSR text is the Court's penchant (not limited, of course, to patent law) in reciting established legal principles as if newly-minted by the Court.  In this case, a close reading of the text reveals the not surprising (and close to mundane) conclusion that things that are obvious can also be obvious to try, and that under limited circumstances (i.e., a limited – "finite" – number of "identified, predictable solutions" that "lead to the anticipated success") things that are "obvious to try" are also obvious.  This is consistent with twenty years of Federal Circuit precedent, as illustrated by two cases: In re O'Farrell (Fed. Cir. 1988) and Pfizer, Inc. v. Apotex, Inc. (Fed. Cir. 2007).

In O'Farrell, the Federal Circuit had this to say about the relationship between obviousness and what is "merely obvious to try":

It is true that this court and its predecessors have repeatedly emphasized that "obvious to try" is not the standard under §103.  However, the meaning of this maxim is sometimes lost.  Any invention that would in fact have been obvious under §103 would also have been, in a sense, obvious to try.  The question is: when is an invention that was obvious to try nevertheless nonobvious?

The admonition that "obvious to try" is not the standard under § 103 has been directed mainly at two kinds of error.  In some cases, what would have been "obvious to try" would have been to vary all parameters or try each of numerous possible choices until one possibly arrived at a successful result, where the prior art gave either no indication of which parameters were critical or no direction as to which of many possible choices is likely to be successful.  In others, what was "obvious to try" was to explore a new technology or general approach that seemed to be a promising field of experimentation, where the prior art gave only general guidance as to the particular form of the claimed invention or how to achieve it.

Obviousness does not require absolute predictability of success.  Indeed, for many inventions that seem quite obvious, there is no absolute predictability of success until the invention is reduced to practice.  There is always at least a possibility of unexpected results, that would then provide an objective basis for showing that the invention, although apparently obvious, was in law nonobvious.  For obviousness under § 103, all that is required is a reasonable expectation of success.  (Citations omitted).

And in Pfizer v. Apotex:

Parties before this court often complain that holdings of obviousness were based on the impermissible "obvious to try" standard, and this court has accordingly struggled to strike a balance between the seemingly conflicting truisms that, under 35 U.S.C. § 103, "obvious to try" is not the proper standard by which to evaluate obviousness, but that, under O'Farrell and other precedent, absolute predictability of success is not required.  Reconciling the two is particularly germane to a situation where, as here, a formulation must be tested by routine procedures to verify its expected properties.  The question becomes then, when the skilled artisan must test, how far does that need for testing go toward supporting a conclusion of non-obviousness?

First, this is not the case where there are "numerous parameters" to try.  Rather, the only parameter to be varied is the anion with which to make the amlodipine acid addition salt.  Although we recognize some degree of unpredictability of salt formation, the mere possibility that some salts may not form does not demand a conclusion that those that do are necessarily non-obvious.  This is especially true here, where (1) as noted above, the skilled artisan had a reasonable (although not guaranteed) expectation that amlodipine besylate would form; (2) Pfizer conceded in prior litigation that the type of salt had no effect on the therapeutic effect of the active ingredient, amlodipine, and was practically interchangeable, and (3) numerous other publications (described above) clearly directed the skilled artisan to a pharmaceutically-acceptable acid addition salt made from benzene sulphonate, including, significantly, the Carabateas patent which taught the besylate acid addition salt form of another dihydropyridine pharmaceutical compound.

Second, this is not the case where the prior art teaches merely to pursue a "general approach that seemed to be a promising field of experimentation" or "gave only general guidance as to the particular form of the claimed invention or how to achieve it."  Here, as admitted by Mr. Davison, in selecting an acid addition salt formulation, one skilled in the art looked to pharmacopoeias and compendia to find a salt that was previously approved by the FDA and used successfully within the pharmaceutical industry.  Berge clearly pointed the skilled artisan to 53 anions that, as of 1974, were pharmaceutically acceptable.  As Dr. Wells' testimony and the Carabateas patent demonstrated, one of ordinary skill in the art was capable of further narrowing that list of 53 anions to a much smaller group, including benzene sulphonate, with a reasonable expectation of success.

Finally, Pfizer protests that a conclusion that amlodipine besylate would have been obvious disregards its "discovery" because it was obtained through the use of trial and error procedures.  While the pharmaceutical industry may be particularly adversely impacted by application of an "obvious to try" analysis, that Pfizer had to verify through testing the expected traits of each acid addition salt is of no consequence because it does not compel a conclusion of non-obviousness here.  In coming to this conclusion, we have not ignored the fact that "[p]atentability shall not be negatived by the manner in which the invention was made."  35 U.S.C. § 103(a).  Nor are we ignorant of the fact that reference to "routine testing" or "routine experimentation" is disfavored.

*  *  *

However, on the particularized facts of this case, consideration of the "routine testing" performed by Pfizer is appropriate because the prior art provided not only the means of creating acid addition salts but also predicted the results, which Pfizer merely had to verify through routine testing.

Thus, while patentability of an invention is not negated by the manner in which it was made, "the converse is equally true: patentability is not imparted where 'the prior art would have suggested to one of ordinary skill in the art that this process should be carried out and would have a reasonable likelihood of success.'"  (Citations omitted).

Taken in the context of established Federal Circuit (and C.C.P.A. precedent), the Supreme Court's latest assertion of the relationship between obviousness and what it "obvious to try" accomplishes nothing other than to create the (false) impression that new law, or a new interpretation of the law, has been made.  Not so:  it remains the case that what is obvious is also frequently "obvious to try," because there are a limited number of parameters to be varied or new technological methods are applied to an old problem.  And what is merely "obvious to try," without more (i.e., the limited number of finite and predictable solutions that lead to a predictably successful result) is frequently nonobvious, particularly when the result is unpredictable or indeed, suprising and unexpected.  Perhaps it would have been better had the Supreme Court heeded the cautionary note sounded by the Federal Circuit in DyStar Textilfarben GmbH v. C.H. Patrick Co. (Fed. Cir. 2006):

Obviousness is a complicated subject requiring sophisticated analysis, and no single case lays out all facets of the legal test.  [There is] danger inherent in focusing on isolated dicta rather than gleaning the law of a particular area from careful reading of the full text of a group of related precedents for all they say that is dispositive and for what they hold.  When parties . . . do not engage in such careful, candid, and complete legal analysis, much confusion about the law arises and, through time, can be compounded.

For additional information and commentary regarding the KSR decision, please see:

• "Like a Penny Saved for a Rainy Day (Albeit, Unintentionally): The Renewed Relevance of 35 U.S.C. § 103(b)," May 13, 2007
• "BIO's Reaction to KSR v. Teleflex," May 8, 2007
• "Implications of the Supreme Court's KSR v. Teleflex Decision for Biotechnology," May 4, 2007
• "The Patent Office Reacts to KSR: A First Look," May 3, 2007
• "KSR Int'l Co. v. Teleflex Inc. (2007)," April 30, 2007
• "Conference & CLE Calendar," which provides an updated list of upcoming KSR CLE's

Comments

[Brad Crawford]

Your analysis suggests the courts may find an invention obvious if it is 1)obvious to try, 2) there are a finite number of predictable solutions, and 3) a predictably successful result.

So, my question is: How does this impact the patentability of single enantiomers?

If a racemate has a desired effect, it is known in the art that one of the isolated enantiomers will likely have superior efficacy, binding, etc., when compared to the other enantiomer (a predictable result, although the identity of the "better" enantiomer cannot be ascertained until both are tested).

It is arguably obvious to try and resolve the enantiomers. The pharmaceutical industry has been doing this for decades.

Since there are only two enantiomers (a finite number of solutions) and one of ordinary skill in the art would know of a finite number of methods, reagents, procedures that are taught by the prior art and are useful for resolving enantiomers (which makes the isolation of any given pair of enantiomers predictable, i.e., the prior art teaches that enantiomers can be separated) are the isolated enantiomers then obvious?

[Kevin E, Noonan]

Exactly. The only way to overcome this conclusion would be 1) if there was no suggestion that one enantiomer was more active than the other, and/or 2) the active enantiomer has MUCH higher (i.e., unpredictably higher) activity than the racemate.

There is also some comfort in that chemistry is inherently much less predictable than the kind of mechanical equivalents in KSR.

Finally, you might argue that you didn't know which enantiomer was active, and the skilled worker did not know how to separate them (i.e., if they were not separated using routine methods known to work for that racemate). Clearly, this argument wouldn't work if you isolated enantiomer A from racemate AB that was related to racemate CD where it was known to be separable into enantiomers C and D (and where one enantiomer had more activity than the other).

Thanks for the comment.