874 F.2d 804
Fed. Cir.
1989

Lead Opinion

OPINION

MAYER, Circuit Judge.

Merck & Co., Inc. sued Biocraft Laboratories, Inc. for infringement of its patent, U.S. Patent No. 3,781,430. The United States District Court for the District of New Jersey held that Biocraft failed to show by clear and convincing evidence that the patent was invalid or unenforceable, and enjoined Biocraft from commercially making, using, and selling the claimed diuretic formulations. Merck & Co., Inc. v. Biocraft Laboratories, Inc., 690 F.Supp. 1376 (D.N.J.1988). Biocraft appeals only the validity aspect of the judgment, arguing that the claimed combinations would have been obvious under 35 U.S.C. § 103. We agree and accordingly reverse the judgment of the district court.

Background

The %30 Patent:

In relevant part, U.S. Patent No. 3,781,-430 (the ’430 patent), assigned to Merck, claims various “Diuretic Formulations.” Inducing diuresis (increased urine excretion) or saluresis (increased electrolyte, particularly sodium, excretion), or both, diuretics are useful in the treatment of cardiovascular and renal diseases. Although the therapeutic effects of diuretics are often lifesaving, many cause hypokalemia, the excessive excretion of potassium ions, a condition manifested by severe muscle weakness and physical exhaustion.

In the formulations claimed in the '430 patent, amiloride hydrochloride (amiloride), a known “potassium conserving” diuretic (induces sodium, but not potassium, excretion), is combined with hydrochlorothiazide, a known “potassium excreting” diuretic (induces both sodium and potassium excretion). As stated in the specification, the objective of co-administration is to reduce the amount of potassium ions eliminated, without reducing the amount of sodium ions eliminated. The inventor also reports that co-administration results in a “medically ... synergistic therapeutic accomplishment” because “more sodium ions are eliminated than would be forecast from a knowledge of the natriuretic [sodium excreting] effects of the individual drugs.”

The ’430 patent contains six claims. The parties agree, however, that all of the claims stand or fall with claims 2 and 3. Claim 1, on which claim 2 depends, and claims 2 and 3 provide as follows:

1. A composition for oral administration comprising amiloride hydrochloride and *806hydrochlorothiazide, wherein the ratio of amiloride hydrochloride to hydrochloroth-iazide ranges from about 1:1 to 1:10 by weight of the composition.
2. A composition according to claim 1 wherein amiloride hydrochloride and hy-drochlorothiazide are combined at a ratio of 1 to 10 by weight.
3. A composition for oral administration which comprises 5 mg. of amiloride hydrochloride and 50 mg. of hydrochloroth-iazide.

The Infringement Suit:

Merck’s suit against Biocraft was prompted by Biocraft’s filing of an abbreviated new drug application (ANDA) with the Food & Drug Administration (FDA) for a generic version of Merck’s amiloride/hy-drochlorothiazide combination, “Moduretic.” The Drug Price Competition and Patent Term Restoration Act of 1984 permits the filing if the generic manufacturer can certify its belief that the patent is invalid or will not be infringed by its proposed manufacture, use or sale of the drug, and notifies the patent owner of the reasons for its belief. 21 U.S.C. § 355(b)(2)(A)(iv) and (3). However, under 35 U.S.C. § 271(e)(2), the filing is a technical act of infringement, so infringement is not an issue in this case. If a patent owner brings an infringement suit, the FDA cannot approve the ANDA unless the patent is declared invalid or not infringed. 21 U.S.C. § 355(c)(3)(C).

At trial, Biocraft argued that the '430 patent was invalid for obviousness under 35 U.S.C. § 103, and unenforceable because of inequitable conduct. The district court ruled against Biocraft on both counts. Bio-craft predicated its section 103 argument on two prior art patents and on prior art relating to spironolactone and triamterene, diuretics that, like amiloride, are potassium conserving. Refreshingly, Biocraft challenges the judgment of the district court only as to obviousness, and restricts its arguments to the teachings of U.S. Patent No. 3,313,813 (the '813 patent), also assigned to Merck.

The '813 patent discloses various (3-ami-no-5,6-disubstituted-pyrazinoyl) guani-dines, one of which is amiloride (claim 11). Per the specification, the claimed compounds are effective diuretic and natriuretic agents. Moreover, the '813 patent teaches that guanidines “selectively enhance the excretion of sodium ions without causing an increase in excretion of potassium ions,” and “are useful in combination with other classes of diuretic agents to prevent the loss of potassium which the other diuretics otherwise would cause to be eliminated.” Hydrochlorothiazide is identified as an example of a potassium excreting diuretic with which the claimed compounds can be combined.

The ’813 patent therefore teaches a genus of which the claims of the ’430 patent are a species. The question addressed in the district court was whether the ’813 patent taught the specific 1:10/5 mg:50 mg, “medically synergistic,” amiloride/hydro-chlorothiazide, combination claimed in the ’430 patent. Biocraft argued that the claimed combination was taught because both amiloride and hydrochlorothiazide were highlighted in the ’813 patent. Moreover, Biocraft pointed out that the Patent and Trademark Office considered the claimed combination prima facie obvious, and allowed the '430 patent only after Merck presented evidence allegedly demonstrating a natriuretic synergism produced by the combination. Asserting that the alleged synergistic effect is not exhibited by the 5 mg:50 mg combination, and that therefore, the effect, even if produced, is not commensurate in scope with the claims, Biocraft argued that neither the purported effect nor the dosage limitations distinguish the claims from the disclosure of the '813 patent. The district court, however, was not persuaded, and held that Biocraft failed to carry its burden of proving the obviousness of the claims. This appeal followed.

Discussion

The district court found that the combination of amiloride and hydrochlorothiazide was disclosed in the '813 patent. 690 F.Supp. at 1383. Further finding that more than 1200 combinations are disclosed, *807and that neither amiloride nor hydrochlo-rothiazide are highlighted as preferred embodiments in the ’813 patent, however, the district court concluded that the combinations claimed in the ’430 patent would merely be “obvious to try,” and therefore not barred by 35 U.S.C. § 103. We do not quarrel with the factual findings of the district court, but we believe its conclusion that obviousness had not been proven is incorrect as a matter of law.

An invention is “obvious to try” “where the prior art [gives] either no indication of which parameters [are] critical or no direction as to which of many possible choices is likely to be successful.” In re O’Farrell, 853 F.2d 894, 903, 7 USPQ2d 1673, 1681 (Fed.Cir.1988). This is not the situation here. The ’813 patent expressly teaches “that when co-administered with other diuretic agents known to enhance the elimination of potassium ions along with sodium ions, the novel pyrazinoylguani-dines of this invention will reduce the excretion of potassium ions and thus overcome this undesirable property of other diuretic agents.” As is apparent, “success” is not dependent upon random variation of numerous parameters. On the contrary, the ’813 patent instructs the artisan that any of the 1200 disclosed combinations will produce a diuretic formulation with desirable sodium and potassium eliminating properties.

That the ’813 patent discloses a multitude of effective combinations does not render any particular formulation less obvious. This is especially true because the claimed composition is used for the identical purpose taught by the prior art. See In re Corkill, 771 F.2d 1496, 1500, 226 USPQ 1005, 1008 (Fed.Cir.1985) (obviousness rejection of claims affirmed in light of prior art teaching that “hydrated zeolites will work” in detergent formulations, even though “the inventors selected the zeolites of the claims from among ‘thousands’ of compounds”); In re Susi, 440 F.2d 442, 445, 169 USPQ 423, 425, 58 CCPA 1074 (1971) (obviousness rejection affirmed where the disclosure of the prior art was “huge, but it undeniably include[d] at least some of the compounds recited in appellant’s generic claims and it is of a class of chemicals to be used for the same purpose as appellant’s additives”).

Merck imputes undue significance to the district court’s finding that neither amiloride nor hydrochlorothiazide are highlighted in the ’813 patent. The description of “specific preferences in connection with a generic formulá” is determinative in an analysis of anticipation under 35 U.S.C. § 102. In re Petering, 301 F.2d 676, 681, 133 USPQ 275, 279, 49 CCPA 993 (1962); see also In re Schaumann, 572 F.2d 312, 315, 316, 197 USPQ 5, 8, 9 (CCPA 1978) (“the disclosure of a chemical genus ... constitute^] a description of a specific compound” within the meaning of section 102(b) where the specific compound falls within the ambit of a “very limited number of compounds”). But in a section 103 inquiry, “the fact that a specific [embodiment] is taught to be preferred is not controlling, since all disclosures of the prior art, including unpreferred embodiments, must be considered.” In re Lamberti, 545 F.2d 747, 750, 192 USPQ 278, 280 (CCPA 1976).

Merck also attributes determinative importance to the district court’s finding that the ’813 patent does not describe “the effects that the combined drug would have on sodium excretion, potassium excretion and ratio of sodium excretion to potassium excretion” that are described in the specification of the ’430 patent, 690 F.Supp. at 1383, nor does it “teach the specific ratios of the combinations” disclosed in the ’430 patent, id. But, the mere absence from the prior art of a teaching or a limitation recited in the patent at issue is insufficient for a conclusion of nonobviousness. Unlike a section 102 defense which requires that a single reference describe each and every element of a claimed invention, Structural Rubber Products Co. v. Park Rubber Co., 749 F.2d 707, 715, 223 USPQ 1264, 1270 (Fed.Cir.1984), “the question under 35 USC 103 is not merely what the references expressly teach but what they would have suggested to one of ordinary skill in the art at the time the invention was made.” In re Lamberti, 545 F.2d at 750, 192 USPQ at *808280. That this distinction was missed is evident from the district court’s erroneous assertion that because the ’813 patent “ ‘suggests,’ ” but does not “ ‘teach’ ” the combinations claimed in the ’430 patent, obviousness had not been proven. 690 F.Supp. at 1383.

As prescribed by section 103, the proper focus of an obviousness inquiry is on whether “the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art....” Merck argues that the dosages and ratios of claims 2 and 3, and the properties of the 5 mg of amiloride and 50 mg of hydrochlo-rothiazide combination patentably distinguish the claims of the ’430 patent from the teaching of the ’813 patent. Relying on evidence allegedly demonstrating that the properties Merck posits are not exhibited by the 5 mg/50 mg combination, Biocraft insists that the claims are invalid under section 103; but this begs the question. Even if Merck’s representations are accepted, the issue remains whether the asserted properties and proportions impart patenta-bility to the claimed inventions, or whether the inventions as a whole would have been obvious to one skilled in the art.

Merck offers several theories to bolster the claimed combinations' purportedly non-obvious properties. In response to the examiner’s rejection of the claims on the ground that the amiloride/hydrochlorothia-zide combination was prima facie obvious, Merck said that the co-administration of the two diuretics achieved a “medically synergistic” result. This apparently convinced the examiner, but it should not have.

“The patent law ‘allows the inventor to be his own lexicographer,'” Loctite Corp. v. Ultraseal Ltd., 781 F.2d 861, 867, 228 USPQ 90, 93 (Fed.Cir.1985) (quoting Autogiro Co. of America v. United States, 384 F.2d 391, 397, 155 USPQ 697, 702, 181 Ct.Cl. 55 (1976)), but he may not compose his own standards of patentability. Synergism is not a requirement of nonobviousness. Gardner v. TEC Sys. Inc., 725 F.2d 1338, 1349, 220 USPQ 777, 786 (Fed.Cir.1984) (in banc). But when an inventor tries to distinguish his claims from the prior art by introducing evidence of unexpected “synergistic” properties, the evidence should at least demonstrate “an effect greater than the sum of the several effects taken separately.” Sakraida v. Ag Pro, Inc., 425 U.S. 273, 282, 96 S.Ct. 1532, 1537, 47 L.Ed.2d 784, 189 USPQ 449, 453 (1976). It is insufficient that, as Merck defines “medical synergism” in the specification of the ’430 patent, the “two drugs react favorably.”

At trial, Merck altered its approach and tried to show that a greater than additive effect, in terms of sodium excretion, is obtained when amiloride and hydrochlo-rothiazide are co-administered. It failed. The district court said only that amiloride and hydrochlorothiazide can be co-administered “without a reduction in the amount of sodium ions that are eliminated,” 690 F.Supp. at 1378, and that the effect of co-administration as set forth in the specification of the ’430 patent was not described in the '813 patent, 690 F.Supp. at 1383. Merck's proposed finding that an unexpectedly strong natriuretic effect occurs was not adopted by the district court. Moreover, on appeal, Merck says only that the evidence at trial “demonstrated that the combination causes more sodium excretion than 50 mg of hydrochlorothiazide alone”. This is quite different from the proposition advanced at trial, and imparts no patentability to the claimed inventions.

Given the prior art teaching that both amiloride and hydrochlorothiazide are natri-uretic, it is to be expected that their co-administration would induce more sodium excretion than would either diuretic alone. See In re Crockett, 279 F.2d 274, 276, 126 USPQ 186, 188, 47 CCPA 1018 (1960) (the “joint use [of magnesium oxide and calcium carbide] is not patentable” where the prior art teaches “that both magnesium oxide and calcium carbide, individually, promote the formation of a nodular structure in cast iron, and it would be natural to suppose that, in combination, they would produce the same effect and would supplement each *809other”). Indeed, the inventor named on both the ’813 and the ’430 patents, so testified. When asked: “Isn’t it also normal when you administer a potassium sparing compound along with hydrochlorothiazide a period of time, that there would be some increase in the amount of sodium excretion?”, he responded: “That is a possibility. That is not an assured consequence.” But, “absolute predictability of success” is not the criterion; “[f]or obviousness under § 103, all that is required is a reasonable expectation of success.” In re O’Farrell, 853 F.2d at 903, 7 USPQ2d at 1681. When further questioned on the point, the inventor indicated that his uncertainty inhered not in the fact that an increase was to be expected, but only in the magnitude of the increase. This, he testified, “Depends on the amount of potassium sparing compound that you have.”

Merck’s further contention that the claimed inventions as a whole would not have been obvious because of the recited dosage limitations is equally unpersuasive. “Normally, it is to be expected that a change in temperature, or in concentration, or in both, would be an unpatentable modification.” In re Aller, 220 F.2d 454, 456, 105 USPQ 233, 235, 42 CCPA 824 (1955). Patentability may be imparted, however, if the results achieved at the designated concentrations are “unexpectedly good.” In re Antonie, 559 F.2d 618, 620, 195 USPQ 6, 8 (CCPA 1977). As discussed above, this is not the situation here. Assuming, as Merck alleges, that “[t]he specific combination of 5 mg of amiloride and 50 mg of hydrochlorothiazide results in no decrease and indeed an increase in sodium excretion over hydrochlorothiazide alone,” this was to be expected from the known natriuretic properties of the two diuretics.

The evidence at trial showed that, though requiring time and care, the experimentation needed to arrive at the claimed dosages was nothing more than routine. “Pat-entability shall not be negatived by the manner in which the invention was made.” 35 U.S.C. § 103. But 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, viewed in light of the prior art.” In re Bow Chemical Co., 837 F.2d 469, 473, 5 USPQ2d 1529, 1531 (Fed.Cir.1988).

This is precisely the situation here. Merck’s expert, the inventor of hydrochlorothiazide, and the witness that the district court found “most closely represents the hypothetical person skilled in the art,” 690 F.Supp. at 1382, testified that the dose response and compatability procedures followed by Merck were those that “all pharmaceutical companies [follow] whenever they determine the appropriate dose; the minimal dose and the appropriate dose.” After describing the animal and human studies that were conducted, he explained: “That is the way it is done not only with amiloride, but with any other drug that you are testing.” Reached by means of routine procedures, and producing only predictable results, the recited dosages therefore do not distinguish the claims of the ’430 patent from the amiloride/hydrochlorothiazide combination that the district court properly found was disclosed in the '813 patent. Cf. United States v. Telectronics, Inc., 857 F.2d 778, 785, 8 USPQ2d 1217, 1223 (Fed.Cir.1988) (specification is enabling in part because those skilled in the art would know how to conduct a dose response study to determine the appropriate amounts to be used).*

Costs

Biocraft moves to tax Merck the full cost of printing the appendix prepared for *810this appeal. Biocraft states that Merck disregarded “the spirit, if not the letter, of Fed.Cir.R. 30(b) by citing in its brief portions of the record not specifically relied upon.” We agree. Our review of the matter shows, for example, that Merck generally cited hundreds of pages of the record in its preliminary introduction of the witnesses, but relied only on a small fraction of those pages in the remainder of its brief; lengthy scholarly articles were reproduced in their entirety but not relied upon by Merck; and a fully reproduced 700-plus page book, though mentioned once by Merck, was similarly not relied upon.

Accordingly, we grant Biocraft’s motion. When assessing costs, the Clerk will tax Merck with the full cost of printing the appendix.

Conclusion

The judgment of the district court is reversed.

REVERSED.

Notes

The district court also referred to evidence of Moduretic’s commercial success. Commercial success is an indication of nonobviousness that must be considered in a patentability analysis, Graham v. John Deere Co., 383 U.S. 1, 17, 86 S.Ct. 684, 693, 15 L.Ed.2d 545, 148 USPQ 459, 467 (1966), but in the circumstances of this case, where it is the only such indication, it is insufficient to render Merck’s claimed invention non-obvious. See W.L. Gore & Assoc., Inc. v. Garlock, Inc., 721 F.2d 1540, 1555, 220 USPQ 303, 314 (Fed.Cir.1983) (objective evidence of nonob-viousness “may in a given case be entitled to more weight or less, depending on its nature and its relationship to the merits of the invention”).



Dissenting Opinion

BISSELL, Circuit Judge,

dissenting.

I would affirm the district court’s decision that the claims at issue in United States Patent No. 3,781,430 (’430) were not shown to have been obvious under 35 U.S. C. § 103 (1982). The district court concluded that a combination of amiloride and hy-drochlorothiazide as claimed in the ’430 patent was merely “obvious to try” based on the prior art. Merck & Co., Inc. v. Biocraft Laboratories, Inc., 690 F.Supp. 1376, 1381 (D.N.J.1988) (citing In re Geiger, 815 F.2d 686, 688, 2 USPQ2d 1276, 1278 (Fed.Cir.1987) (rejecting “obvious to try” as the standard for determining obviousness)).

Obviousness, a question of law, is based on underlying factual inquiries. See Uniroyal, Inc. v. Rudkin-Wiley Corp., 837 F.2d 1044, 1050, 5 USPQ2d 1434, 1438 (Fed.Cir.), cert. denied, — U.S. -, 109 S.Ct. 75, 102 L.Ed.2d 51 (1988). The parties agree that the closest prior art to the ’430 patent is United States Patent No. 3,313,-813 (’813). Additionally, the majority opinion does not “quarrel with the factual findings of the district court.”

In pertinent part the district court stated:

The ’813 patent discloses appropriately 120 pyrazinoylguanidine compounds. There is no indication in the patent that amiloride is the preferred compound. Thus, a person skilled in the art, would not be taught by the ’813 patent that amiloride and hydrochlorothiazide should be combined to produce the effects as set forth in the ’430 patent. The ’813 patent does not describe or forecast the effects that the combined drug would have on sodium excretion, potassium excretion and ratio of sodium excretion to potassium excretion. A composition must be assessed for obviousness only after consideration of its chemical structure as well as its pharmaceutical and biological properties. In short, while the ’813 patent “suggests” exploration of combinations of potassium sparing guanidine with potassium losing diuretics, such as hydrochlorothiazide, it does not “teach” the combinations claims in the ’430 patent.

Nor does it teach the specific ratios of the combinations as outlined in Claims 2, 3, 5 and 6 of the ’430 patent. Those claims are sufficiently non-obvious so as to sustain the validity of the patent. Since the ’813 patent discloses in excess of 120 pyrazinoylguanidine and 10 possible potassium losing diuretics, there are more than 1200 possible combinations, of which the combination of Amiloride and Hydrochlorothiazide would be one. And there is no indication in the ’813 patent as to the preference for amiloride as opposed to any other pyrazinoylguani-dine compound, nor hydroclorothiozide [sic] as opposed to other diuretics. Nor is there any hint as to the proportions that would be contained in the mixture. Again, while the ’813 patent might titillate one’s desire to experiment, it clearly does not teach and thereby make obvious the claims as outlined in Claims 2, 3, 5 and 6 of the ’430 patent.

Merck, 690 F.Supp. at 1383-84.

The majority opinion incorrectly states that the district court found “the combination of amiloride and hydrochlorothiazide was disclosed in the ’813 patent.” The *811district court, in fact, found that the ’813 patent disclosed “in excess of 120 pyrazi-noylguanidine and 10 possible potassium losing diuretics, [resulting in] more than 1200 possible combinations, of which the combination of Amiloride and Hydrochlo-rothiazide would be one.” Id. The ’813 patent does not disclose a diuretic composition containing amiloride and hydrochlo-rothiazide.

After examining the scope and content of the prior art, the district court found that the ’813 patent does not (1) teach one of ordinary skill in the art a preference for amiloride, hydrochlorothiazide, or their combination, (2) “describe or forecast the effects that the combined drug would have on sodium excretion, potassium excretion and ratio of sodium excretion to potassium excretion,” and (3) teach the specific ratio limitations of the claims in suit. Id. Additionally, the district court determined that the ’813 patent only suggests “exploring” combinations of potassium sparing and potassium losing diuretics, and that the prior art does not teach one of ordinary skill what is claimed in the ’430 patent. Id. I agree. The claimed invention would not have been obvious in view of prior art that does no more than suggest experimenting with over 1200 combinations to come up with the right one.

The meaning to one of ordinary skill in the art of the ’813 patent was extensively explored at trial. Based upon this testimony, the district court found that Merck’s expert, the inventor of hydrochlorothiazide, “most closely represented] the hypothetical person skilled in the art.” Id. at 1382. After reviewing the prior art, Merck’s expert testified that in 1966 the combination of amiloride and hydrochlorothiazide would not have been obvious, and if combined, he would not have expected the results achieved.

In examining secondary considerations, the district court found “uncontroverted evidence” of commercial success, id. at 1385 n. 4, but made no express findings on unexpected results. Yet the majority, after extended discussion, finds no unexpected results. Absent more definitive findings by the district court, the basis for the majority’s statement escapes me.

Hindsight is not the standard for determining obviousness. See, e.g., Uniroyal, 837 F.2d at 1050-51, 5 USPQ2d at 1438. Bits and pieces of the invention claimed in the ’430 patent can be pointed to in the prior art, but only if one is armed with hindsight knowledge. See id. (stating that hindsight reconstruction using the claimed invention as a blueprint is improper). The majority’s opinion does not disturb the district court’s factual findings. Based on these findings, I cannot agree with the majority’s conclusion that the claims of the ’430 patent would have been obvious. Accordingly, I would affirm the district court’s judgment.