In Re Raymond C. Grabiak

PAULINE NEWMAN, Circuit Judge. Raymond C. Grabiak et al. appeal from the decision of the Patent and Trademark Office Board of Appeals sustaining the rejection of claims 1 through 34, all of the claims of patent application Serial No. 168,-959, filed July 17, 1980 for “2-Chloro-4-Trifluoromethyl Thiazolecarbothioic Acids Useful As Herbicidal Safeners”, as unpatentable under 35 U.S.C. § 103. We conclude that the PTO has not presented a prima facie case of unpatentability, and on this basis we reverse the decision of the Board. The Invention The claimed invention relates to a class of chemical compounds having utility as herbicidal safeners. Safeners, sometimes called antidotes, are used to protect growing crops from damage that may be caused by the application of herbicides to control undesired plants. The claimed compounds, useful as safeners against acetanilide herbicides, are certain thiazole thiocarboxy *730 lates as shown in Claim 1, the broadest claim: 1. A compound of the formula wherein R is Cl-5 alkyl, phenyl or benzyl. Other claims are directed to various species, to herbicidal mixtures containing these compounds, and to various methods of use of these compounds. Grabiak has not argued the claims separately, and we do not so consider them. The Rejection The claims stand rejected as obvious from Howe et al. U.S. Patent No. 4,199,-506. Also relied on are Bollinger U.S. Patent No. 4,317,310 and R. Conant & A. Blatt, The Chemistry of Organic Compounds 342-43 (3d ed. 1947), an organic chemistry textbook. Howe describes a family of chemical compounds having utility as safeners for acetanilide herbicides, consisting of thiazole carboxylic and thiazole carboxamide compounds of the general formula: In the Howe disclosure R, R', n, and X, are broadly defined, the breadth of which is not pertinent to this issue. Very pertinent is the disclosure in Howe of the following specific compound: This compound differs from those claimed by Grabiak only by the presence in Grabiak of a sulfur atom instead of a particular oxygen atom in the ester moiety, a difference which the examiner asserted would have been, without more, obvious. The examiner cited the Bollinger reference as showing the interchangeability of oxygen and sulfur in compounds having safening properties. Bollinger shows, as safeners for thiocarbamate and acetanilide herbicides, a class of 2-imino derivatives of 1,3-oxathioles and 1,3-dithioles. The examiner pointed to the 1,3-oxathiole/dithiole ring fragment: wherein Z is defined as either oxygen or sulfur, as support for the conclusion that it would have been obvious to exchange a sulfur atom for an oxygen atom in the Howe compounds. The Board agreed. *731 On reconsideration, the Board in a split decision affirmed the rejection, citing In re Fancher, 410 F.2d 813, 161 USPQ 613 (CCPA 1969) and In re Albrecht, 579 F.2d 92, 198 USPQ 208 (CCPA 1978) for the proposition that oxygen and sulfur are well known to be interchangeable. To “reiterate that the close analogy between sulfur and oxygen isologs is well known,” the Board referred to Conant & Blatt’s discussion of the general similarities between simple sulfur and oxygen compounds. One member of the Board dissented, stating his belief that the compounds disclosed in Bollinger are “too remote to those claimed” to suggest substitution of sulfur for oxygen at a particular place in the Howe compounds. The Argument Grabiak presented no evidence that his safener compounds have unobvious properties as compared with Howe’s safener compounds, and stated plainly that they do not. Grabiak’s argument is, in sum, that (1) in the field of biological activity, it is not predictable whether chemical compounds that have an apparent structural similarity will also have similar biological properties; (2) biological properties cannot be predicted; they must be determined by experimentation; (3) therefore mere structural similarity is inadequate to present a prima facie case of obviousness; and (4) more is required, such as suggestion in the prior art (a) that the structural modification should be made and (b) that the modified compound will exhibit the biological behavior of the prior art compound. Grabiak argues that Howe does not teach that one of the oxygens in the Howe carboxylate group could be replaced with sulfur to produce safeners for acetanilide herbicides, and that Bollinger and Conant & Blatt do not cure this deficiency because Bollinger is dealing with a quite different part-of a quite different molecule, and the Conant & Blatt text refers only to simple structures and chemical, not biological, properties; and in any event that safening activity is, like all biological behavior, unpredictable. Grabiak asserts that the teachings of Howe with Bollinger and Conant & Blatt are insufficient to establish prima facie obviousness, in that there is no motive in the cited art to make the modification required to arrive at appellants’ compounds. Analysis I. When chemical compounds have “very close” structural similarities and similar utilities, without more a prima facie, ease may be made. See for example In re Wilder, 563 F.2d 457, 195 USPQ 426 (CCPA 1977) (adjacent homologues and structural isomers); In re May, 574 F.2d 1082, 197 USPQ 601 (CCPA 1978) (steroisomers); In re Hoch, 428 F.2d 1341, 166 USPQ 406 (CCPA 1970) (acid and ethyl ester). When such “close” structural similarity to prior art compounds is shown, in accordance with these precedents the burden of coming forward shifts to the applicant, and evidence affirmatively supporting unobviousness is required. Analysis of those circumstances in which a prima facie case has or has not been made in view of the degree of structural similarity or dissimilarity, or the presence or absence of similar utility between the prior art compound and that of the applicant, has inspired generations of applicants, courts, and scholars. Upon review of this history, we have concluded that generalization should be avoided insofar as specific chemical structures are alleged to be prima facie obvious one from the other. Although we do not accept Grabiak’s argument that when biological activity is involved there can be no presumption (i.e. no prima facie case) of obviousness, in the case before us there must be adequate *732 support in the ■ prior art for the ester/thioester change in structure, in order to complete the PTO’s prima facie case and shift the burden of going forward to the applicant. The Bollinger teaching of various hetero-cyclic rings containing either two sulfur atoms or one oxygen and one sulfur atom, rings which are unlike any part of the Howe molecule, does not suggest the interchangeability of sulfur for oxygen in the ester moiety of the Howe molecule. (Grabiak also analyzes the Bollinger disclosure as showing “dramatic decreases in safener activity when replacing oxygen with sulfur”.) Conant & Blatt’s brief discussion that “simple sulfur compounds” have properties similar to simple oxygen compounds does not purport to apply to complex organic molecules. Nor do the Fancher and Albrecht cases remedy these deficiencies, for in each of those cases the sulfur/oxygen interchange was in a heterocyclic ring common to both the prior art compounds and the applicant’s compounds. We repeat the statement of In re Berg el, 292 F.2d 955, 956-57, 130 USPQ 206, 208 (CCPA 1961), that: The mere fact that it is possible to find two isolated disclosures which might be combined in such a way to produce a new compound does not necessarily render such production obvious unless the art also contains something to suggest the desirability of the proposed combination. [Emphasis in original] The PTO cited no pertinent reference showing or suggesting to one of ordinary skill in the art the change of a thioester for an ester group. In the absence of such reference, there is inadequate support for the PTO’s position that this modification would prima facie have been obvious. II. The Solicitor contends that the sulfur in Grabiak that replaced the oxygen in Howe occurs in a portion of the molecule that is not significant to safener activity, as further argument that Grabiak’s compounds would have been obvious from Howe’s compounds. To support this argument the Solicitor refers to the statement in Howe that the carboxylic moiety may include the acid and salts thereof, acid chlorides, amides, and esters. From this the Solicitor argues that the nature of this moiety “would not be expected to impart or contribute to the safening utility”, and therefore that the replacement of' Howe’s ester with Grabiak’s thioester would have been obvious. This argument is lacking in a critical element: adequate support in the prior art. Howe does not state that the carboxylic segment of his molecule is not significant to its biological properties, and no other support is invoked. We appreciate that the PTO lacks the possibility of experimental verification of this theory; but absent an initial prima facie case, we do not think the burden of disproving this theory is shifted to Grabiak. Nor do we judicially accept a theory that appears to require the general assumption that sulfur is not significant to biological behavior. Grabiak argues further that the PTO’s position that the identity of the carboxylic component is not material cannot apply here because safening activity can not be predicted from chemical structure. Grabiak asserts that the efficacy of any compound for safening depends on variables including the type of herbicide compound, the type of weed to be controlled, the type of crop to be protected and the safener compound itself. Grabiak cites data from Howe which he states show that a “compound, which safens one herbicide used to control barnyard grass in the presence of corn crop, is totally ineffective to safen that same herbicide to control barnyard grass in the presence of rice.” Grabiak also cites data from Bollinger to support Grabiak’s position that “safening activity even for closely similar homologues does not vary predictably.” *733 In response, the Solicitor argues that it is not “necessarily true” that safening activity is not predictable from the structure of the compound. Evidence for this statement is seen by the Solicitor in Grabiak’s compounds themselves, which are admitted to have the same safening activity as those of Howe. However, Grabiak’s disclosure may not be used to fill the gaps in the prior art. If evidence of similar biological properties between -C(0)0R and -C(0)SR groups is to be relied upon, it must come from the prior art. The PTO produced no such evidence. Instead, the Board held that “it is not inconceivable to substitute [sulfur for oxygen] to obtain compounds having the same expected properties.” We agree that it is not inconceivable. The standard, however, is whether it would have been obvious in terms of section 103. In the absence of adequate support, we conclude that this argument does not perfect the PTO’s prima facie case. III. We have considered the decisions on which the PTO relies. In In re Payne, 606 F.2d 303, 203 USPQ 245 (CCPA 1979) there was prior art well supporting the PTO’s prima facie case. In In re Susi, 440 F.2d 442, 169 USPQ 423 (CCPA 1971) the difference from the prior art compound was a hydroxyl group, a difference that the applicant conceded was “of little importance”. In In re Doebel, 461 F.2d 823, 174 USPQ 158 (CCPA 1972), the court stated that “the claimed compound is a homologue”, and a prima facie case was held to have been made. None of these cases requires the result that a thioester derivative be deemed prima facie obvious from the corresponding ester in the absence of prior art on this point. Conclusion On the record before us, we conclude that the PTO did not establish a prima facie case of obviousness, and thus did not shift to Grabiak the burden of coming forward with evidence of unexpected results. REVERSED.