This appeal is from the decision of the United States Patent and Trademark Office (PTO) Board of Appeals (board) affirming the rejection of claims 1-5 in appellants’ application, serial No. 461,011, filed April 15,1974, for “Androst-4-en-19-ols,” under 35 U.S.C. § 103 as obvious from the teachings of the prior art references. We affirm the rejection of claims 1 and 5. We reverse the rejection of claims 2 — 4.
The Invention
Appellants disclose that certain allegedly novel derivatives of androst-4-en-19-ol steroid compounds have unexpected psychological and physiological properties. Specifically, these derivatives are said to enhance diminished libido and related psychic attitudes in man and other primates without demonstrating any overt androgenic or es-trogenic response upon secondary sex characteristics.
Illustrative claim 1 reads as follows (emphasis and position numbers ours):
1. An androst-4-en-19-ol having the formula
wherein
Ri is selected from the group consisting of H2, oxo and H(OR2) with the proviso that when Rg, R7, Rs and R9 are all hydrogen and Ri is H2 or oxo and R2 is hydrogen or acyl, then R4 or R5 must be selected from the group consisting of lower alkyl having from 1 to 3 carbon atoms, trialkylsilyl in which the alkyl group has from 1 to 5 carbon atoms, triphenylsilyl, 2-tetrahydropyranyl, 4-tetrahydropyranyl, 1-cycloalkenyl having from 5 to 7 carbon atoms, 1-me-thoxycycloalkyl and 1-ethoxycycloalkyl in which the cycloalkyl group has from 5 to 7 carbon atoms;
R2i R4 and R5 are each selected from the group consisting of hydrogen, acyl having from 1 to 12 carbon atoms, lower alkyl having from 1 to 3 carbon atoms, trialkylsilyl in which the alkyl group has from 1 to 5 carbon atoms, triphe-nylsilyl, 2-tetrahydropyranyl, 4-tetrahy-dropyranyl, 1-cycloalkenyl having from *4885 to 7 carbon atoms, 1-methoxycycloal-kyl and 1-ethoxycycloalkyl in which the cycloalkyl group has from 5 to 7 carbon atoms, with the proviso that when R4 is 2-tetrahydropyranyl and Rg, R7, Rs, and Rg are all hydrogen, then Rg and Rg cannot both be hydrogen and Ri cannot be oxo;
Rg is selected from the group consisting of hydrogen, lower alkyl having from 1 to 6 carbon atoms, lower alkenyl having from 2 to 6 carbon atoms and lower alkynyl having from 2 to 6 carbon atoms, and Rg and OR4 when taken together is oxo;
Rg, R7, Rg and Rg are hydrogen and methyl with the proviso that when Ri is oxo, R4 is hydrogen, acyl or when taken, together with Rg is oxo, Rg is hydrogen and Rg is methyl, then Rg, R7 and Rg cannot all be hydrogen at the same time.
The References
Mihina 2,819,276 Jan. 7, 1958
Bowers et al. 3,278,528 Oct. 11, 1966
Applezweig, Steroid Drugs, McGraw-Hill Book Company, Inc., New York, Page 273 (1962).
Mihina discloses 3,19-dihydroxy steroids. Particularly related is androst-4-en-3,19-diol-17-one which has the following structure.
The above compound is an intermediate which is converted to a 19-norandrost-4-en-3-one derivative. Although no properties are described for the 3,19 dihydroxy steroids, the resulting derivatives are stated to have anabolic and hypotensive action.
The Bowers et al. reference relates to 3,17,19 substituted androst-4-enes of the following formula:
OR
In the above formulae, * * * R, R2, and R3 each represent hydrogen or a hydrocarbon carboxylic acyl group of less than 12 carbon atoms; R1 represents hydrogen or a lower hydrocarbon residue of up to 6 carbon atoms * * *.
These steroids are stated to be anabolic-androgenic agents having anti-estrogenic, anti-gonadotrophic, anti-fibrillatory and appetite stimulating properties.
The Applezweig reference describes the therapeutic applications of various steroid drugs. Androst-5-en-3-ol-17-one is said to be only weakly androgenic and capable of producing improved sexual behavior.
Represented by the figure below, this drug also allegedly produces altered behavioral patterns and mental attitudes.
The examiner stated that the “close relationship between ethers and alcohols” renders the claimed ether-substituted androst-4-enes to be structurally obvious from the hydroxy-substituted androst-4-enes of the prior art. Since a prima facie case of structural obviousness was said to be present, the examiner required a demonstration of unexpected properties. He found no such showing based on comparative date to have been made.
The Board
In a first opinion of June 20, 1977, the board affirmed the rejection of claims 1-5, *489stating that the close structural relationship between the prior art androstenes and the claimed compounds establishes a prima fa-cie case of obviousness.
Appellants’ position, that no prima facie case had been established, was held contrary to the prevailing interpretation of the law as set forth in In re Hoch, 428 F.2d 1341, 57 CCPA 1992, 166 USPQ 406 (1970). Furthermore, the board held that the alleged unexpected libido-enhancing properties of the present compounds do not overcome the prima facie case. The board stated that under Hoch, proof of actual unexpected differences in the properties is required to rebut a prima facie case. Appellants’ only showing was said to reside in Example 66 of the instant specification. In that example, tests with castrated rats compared the effects of 19-hydroxy-testoster-one with those of testosterone. Because none of these steroids is encompassed by the present claims, appellants’ comparison was held meaningless and, therefore, insufficient as rebuttal evidence.
The Applezweig reference was first cited by the board, the board treating it as a new ground of rejection if appellant desired to so regard it. The stated purpose of this reference was to demonstrate “a reasonable expectation that the related androstenes would manifest a similar activity.” Appellants were held to be wrong in requiring the prior art to disclose an exact utility.
Appellants’ Request for Reconsideration
Appellants requested reconsideration by the board. The gist of their argument is that the claimed androst-4-enes and the Ap-plezweig androst-5-enes are totally separate and unrelated classes of steroids. Appellants stated:
Thus, for example, the compounds claimed contain an allylic alcohol in the 3-position which causes the molecule to be more readily dehydrated, more readily substituted at the 3-position and to more readily undergo oxidation at this position. Oxidation of the compounds claimed with manganese dioxide, for example, will form the 3-ketone, whereas the prior art compound will remain unaffected.
The above dissimilar chemical properties were said to lead one away from predicting similar biological properties.
Three affidavits were submitted by appellants in support of the above argument. One of the affiants was Mr. Norman Ap-plezweig, author of the reference in question. All of the affidavits essentially state the following:
I. Applezweig “is totally devoid of any disclosure or suggestion whatsoever * * * ” regarding the androst-4-en-19-ols of claim 1;
II. androst-5-en-3pol-17-one is a “chemically separate and distinct entity”; and
III. “subtle structural differences in the steroid art can result in marked differences in biological properties * * * »
Therefore, the affiants concluded that the properties of the specific androst-5-ene of Applezweig cannot be predicted for the claimed compounds.
The board referred appellants’ request to the examiner.
Examiner’s Answer
In a second Answer, the examiner additionally supported his view that a close relationship exists between steroidal alcohols and ethers. Alkoxylation of steroidal alcohols is said to be a modification well known to skilled chemists.1 Furthermore, he said, the utilities of ethers would be obvious from the prior art disclosures.
Reconsideration by the Board
The board again affirmed the rejection by the examiner. Applezweig was said to be cited merely to show that the supposedly unexpected properties of the claimed steroids were known for at least one steroid of *490the androstene series. It was reiterated that the close structural relationship of the prior art establishes a prima facie case of obviousness. The affidavits were held not to effect the establishment of such a showing. The board recognized the qualifications of the experts, but each affidavit, it said, merely stated a conclusion without supporting reasons.
Appellants’ Arguments
Appellants argue that the examiner failed to establish a prima facie case of structural obviousness because none of the references disclose or suggest an ether group located anywhere in steroid molecules.2 Therefore, one would not equate the expected properties of ether-substituted androst-4-enes with the properties of alcohol or hydroxy-substituted androst-4-enes.3
Furthermore, appellants contend that one cannot look to Applezweig to help establish a prima facie obviousness, that the PTO has issued numerous patents directed either to androst-4-enes or androst-5-enes, that three experts state, in affidavits said to have probative value, that the above steroids are totally distinct, and that only a hindsight reconstruction would permit the grouping of the properties of these compounds.
Essentially, the reasoning of the board is said by appellants to require two separate inventions. In U.S. Patent 4,087,524, appellants claimed the use of a class of androst-4en-19-ols which includes both known and novel compounds for heretofore unexpected properties.4 In claiming the novel compounds themselves, however, appellants complain that they must now show unexpected differences within said class solely based on a close structural relationship.
OPINION
The issues are, first, whether or not the androst-4-en-3,19-diols of the prior art render prima facie obvious the claimed 1,4,6, or 7 methyl substituted counterparts and the 17 and/or 19 ether-substituted androst-4-enes; second, if there is a prima facie case, have appellants demonstrated unexpected differences in the properties of any of the claimed compounds which are structurally obvious.
With respect to the first issue, we show below in tabular form the structural differences between the prior art references and the appealed claims for a base androst-4-ene compound.
Substituted Position
3 17 19 Other Positions
Bowers -OH -OH -OH
Claims 1 -OH -OH -OH and 5 —CH3 in 1, 4, 6, or 7 position
Mihina -OH =0 -OH
Claim 2 -OH -OR -OH
Claim 3 -OH =0 -OR
Claim 4 -OH -OR -OR
We hold that claims 1 and 5 would have been prima facie obvious. When the claim provisos are inoperative, the only difference between these claims and the prior art is a methyl substituent. This does not mean that such a substitution on a steroid is per se obvious. Rather, several factors in the prior art compounds suggest structural obviousness in the claimed steroids. An identical base steroid structure (a 1, 2 cyclopen-*491tanophenanthrene ring system) is present in both. The utilities are analogous in their physiological and psychological responses. Appellants admit that androgens are known to affect the libido.5 More importantly, the substituents and their respective 3,17, and 19 active sites are identical. Given these similarities, a skilled steroid chemist at the time the invention was made would not have reason to expect that adding a methyl substituent in an additional position would significantly alter the analogous properties of the references.
The affidavits regarding the use of the Applezweig reference as prior art are of limited probative value. The board correctly states that mere conclusions are set forth therein without cogent, supporting reasons. None of the affiants state any facts which differentiate the properties of androst-4-enes from androst-5-enes. If, as affiants stated, “no reasonable interpretation” can equate the two, affiants should have submitted factual substantiation of that contention. The affidavits, however, fail to offer the bases upon which their conclusions are built. In re Payne, 606 F.2d 303, 315, 203 USPQ 245, 246 (Cust. & Pat.App.1979).
Appellants have also made no showing of unexpected differences in properties. Therefore, the prima facie showing that the compounds encompassed by claims 1 and 5 would have been obvious has not been rebutted.
The remaining claims, 2-4, would not have been obvious unless an equivalence is established between steroidal alcohols and ethers. The references do not support equivalence, despite the examiner’s contentions. Appellants correctly state that:
None of the references disclose or suggest a compound containing an ether group located anywhere in the steroid molecule. The Patent and Trademark Office has cited no art to demonstrate that there is an equivalence between the 17 and/or 19-alcohols and their ethers.
Therefore, the examiner has not shown why one skilled in this art would have replaced a hydroxyl group in the 3,17, and/or 19 position with an ether, to obtain the claimed compounds.
In view of the foregoing, the decision of the board regarding claims 1 and 5 is affirmed. The decision of the board regarding the rejection of claims 2-4 is reversed.
MODIFIED.