RADER, Circuit Judge.
On a motion for judgment as a matter of law (JMOL), the United States District Court for the Southern District of New York upheld the validity of claim 4 of U.S. Patent No. 4,283,408 (the ’408 patent) in favor of Yamanouchi Pharmaceutical Co., Ltd. and Merck & Co., Inc. (collectively, Yamanouchi).
See Yamanouchi Pharm. Co. v. Danbury Pharmacol, Inc.,
21 F.Supp.2d 366, 370, 48 USPQ2d 1741, 1744 (S.D.N.Y.1998). The district court also found that defendants Danbury Pharmacal, Inc. (Danbury), Schein Pharmaceutical, Inc. (Schein), and Marsam Pharmaceuticals, Inc. (Marsam) willfully infringed the ’408 patent and awarded attorney fees to Yamanouchi.
See id.
at 378. Because the district court correctly upheld the validity of the ’408 patent and did not abuse its discretion in awarding attorney fees, this court affirms.
I.
The ’408 patent, issued to Yamanouchi on August 11, 1981, relates to inhibitors of gastric acid secretion. Claim 4 of the ’408 patent, the only claim at issue, claims fam-otidine for treating heartburn and ulcers. Famotidine belongs to a class of compounds known as histamine2 antagonists (H2 antagonists), which inhibit production of stomach acid. As Figure 1 illustrates, the general chemical structure of H2 antagonists includes a “substituted heterocy-cle” group, a “alkyl containing” chain (called a “bridge”), and a “polar tail,” connected in that order:
[[Image here]]
During the 1960s and 70s, drug manufacturers searched for H2 antagonists with improved pharmacological properties, including low toxicity, high activity, and lack of side effects. Research revealed hundreds of thousands of potential compounds. Indeed, pharmaceutical companies synthesized more than 11,000 H2 antagonist compounds.
See Yamanouchi,
21 F.Supp.2d at 371. Very rarely, however, were these compounds pharmacologically suitable H2 antagonists. Notable failures include tioti-dine, which caused cancer in rats; buri-mamide, which was ineffective for oral dosing; metiamide, which caused white blood cell loss; lupitidine, which caused pre-can-cerous lesions in rats; and oxmetidine, which caused hepatitis.
Of the 11,000 candidates for suitable compounds, fewer than fifty showed enough promise to warrant human clinical trials. Ultimately, the FDA approved only
four for consumer use: cimetidine,
raniti-dine,
famotidine,
and nizatidine.
Famo-tidine, the claimed compound at issue, has been extremely successful. In 1996, for example, prescription sales of famotidine in the United States alone reached over 690 million dollars.
Danbury is a subsidiary of Schein, which produces and markets generic drugs. In January 1997, Danbury filed an Abbreviated New Drug Application (ANDA) with the Food and Drug Administration (FDA) seeking approval to market generic famoti-dine. Under the ANDA procedure, an applicant seeks FDA approval to market a generic drug. The Hatch-Waxman Act, also known as The Drug Price Competition and Patent Term Restoration Act. Pub.L. No. 98-417, 98 Stat. 1585 (1984) (codified as amended at 21 U.S.C. § 355 and 35 U.S.C. § 271(e) (1994)), amended the Federal Food, Drug, and Cosmetic Act (FDCA), Pub.L. No. 52-675, 52 Stat. 1040 (1938) (codified as amended at 21 U.S.C. §§ 301-397 (1994)), to permit filing of an ANDA to expedite FDA approval of a generic version of a drug previously approved by the FDA.
See, e.g., Bayer AG v. Elan Pharm. Research Corp.,
212 F.3d 1241, 1244, 54 USPQ2d 1711, 1712 (Fed.Cir.2000).
Under the FDCA, an ANDA filer must certify one of the following four statements concerning the previously approved drug: it is not patented (paragraph I certification), its patent has expired (paragraph II certification), its patent soon will expire on a specified date (paragraph III certification), or its patent “is invalid or will not be infringed by the manufacture, use, or sale of the new drug” covered by the ANDA (paragraph IV certification).
See
21 U.S.C. § 355(j)(2)(A)(vii)(I)-(IV). To obtain approval of an ANDA, the FDCA requires only that the generic drug is the “bioequivalent” of the previously approved drug.
See
21 U.S.C: § 355(j)(2)(A)(iv).
In Danbury’s ANDA for famotidine, Danbury made a paragraph IV certification that claim 4 of the ’408 patent is invalid.
See
21 U.S.C. § 355(j)(2)(A)(vii)(IV). As the statute requires, Danbury, on March 26, 1997, sent Yamanouchi a Patent Certification Notice Letter. This certification letter informed Yamanouchi of Danbury’s paragraph IV ANDA filing. Accompanying the certification letter were affidavits from Drs. Bernard Loev and John K. Siepler supporting Danbury’s invalidity certification. The Notice Letter contained, as the statute requires, an analysis of the prior art and the reasons for the asserted invalidity.
Within forty-five days of receiving the certification letter, Yamanouchi filed suit against Danbury alleging infringement of the ’408 patent under 35 U.S.C. § 271(e)(2)(A), and willful infringement under 35 U.S.C. § 285 (1994).
See
35 U.S.C. § 271(e)(4). During this period, Marsam filed a number of paragraph IV ANDAs seeking FDA approval to market injecta-ble versions of famotidine. Yamanouchi then filed suit against Marsam, and the two suits were consolidated (Danbury, Schein, and Marsam are hereinafter collectively referred to as Danbury). The parties agreed to a bench trial.
After Danbury presented its last witness on obviousness, Yamanouchi moved for JMOL under Fed.R.Civ.P. 52(c) (Rule 52(c)). With that motion, Yamanouchi argued that Danbury had not shown by clear and convincing evidence that claim 4 of the ’408 patent would have been obvious at the time of invention. The district court granted Yamanouchi’s JMOL motion.
See Yamanouchi,
21 F.Supp.2d at 370. Specifically, the district court found that Dan-bury had not shown any motivation to combine selected portions of various prior art compounds to create the specific compound famotidine and to obtain its extraor
dinary properties.
See id.
at 373. The district court characterized Danbury’s case for obviousness as largely hindsight, speculation, and argument without an adequate foundation.
See id.
at 370, 373, 376. Based on those findings, the district court determined that Danbury willfully infringed the ’408 patent. The district court thus found the case “exceptional” and awarded attorney fees and costs to Yamanouchi.
See id.
at 378.
II.
To grant a JMOL under Rule 62(c), a district judge must weigh the evidence and resolve credibility.
See
Fed.R.Civ.P. 52(a) and (c);
Lemelson v. United States,
752 F.2d 1538, 1547, 224 USPQ 526, 530-31 (Fed.Cir.1985). Therefore, this court reviews the district court’s JMOL findings as if entered at the conclusion of all the evidence.
See Lemelson,
752 F.2d at 1547; Woods
v. North Am. Rockwell Corp.,
480 F.2d 644, 645-46 (10th Cir.1973). This court reviews the conclusion on obviousness, a question of law, without deference, and the underlying findings of fact for clear error.
See Univ. of Colo. Found., Inc. v. Am. Cyanamid Co.,
196 F.3d 1366, 1370, 52 USPQ2d 1801, 1803 (Fed.Cir.1999).
A.
Obviousness rests on several critical factual underpinnings: (1) the scope and content of the prior art, (2) the differences between the prior art and the claimed invention, (3) the level of skill in the art, and (4) the objective indicia of nonobviousness.
See Panduit Corp. v. Dennison Mfg. Co.,
810 F.2d 1561, 1566-67, 1 USPQ2d 1593, 1595-96 (Fed.Cir.1987);
Graham v. John Deere Co.,
383 U.S. 1, 17, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966). For a chemical compound, a prima facie case of obviousness requires “structural similarity between claimed and prior art subject matter ... where the prior art gives reason or motivation to make the claimed compositions.”
In re Dillon,
919 F.2d 688, 692, 16 USPQ2d 1897, 1901 (Fed.Cir.1990)
(en banc).
“[A] reasonable expectation of success, not absolute predictability” supports a conclusion of obviousness.
In re Longi,
759 F.2d 887, 896, 225 USPQ 645, 651-52 (Fed.Cir.1985).
As noted earlier, the district court discerned that Danbury had not proven any motivation to combine prior art references to produce the claimed invention. This court has recently reemphasized the importance of the motivation to combine:
As this court has stated, “virtually all [inventions] are combinations of old elements.” Therefore, an examiner [or accused infringer] may often find every element of a claimed invention in the prior art. If identification of each claimed element in the prior art were sufficient to negate patentability, very few patents would ever issue. Furthermore, rejecting patents solely by finding prior art corollaries for the claimed elements would permit an examiner [or accused infringer] to use the claimed invention itself as a blueprint for piecing together elements in the prior art to defeat the patentability of the claimed invention.
... To counter this potential weakness in the obviousness construct, the suggestion to combine requirement stands as a critical safeguard against hindsight analysis and rote application of the legal test for obviousness.
In re Rouffet,
149 F.3d 1350, 1357-58, 47 USPQ2d 1453, 1457 (Fed.Cir.1998) (internal citations omitted).
At the heart of this validity dispute is whether one of skill in this art would have found motivation to combine pieces from one compound in a prior art patent with a piece of another compound in the second prior art patent through a series of manipulations. According to Danbury, one of skill in the art would have considered it obvious to select the example 44 compound from Yamanouchi’s U.S. Patent No. 4,252,-819 (the ’819 patent) and tiotidine from the ’378 patent to use as leads for making
famotidine. These compounds, respectively, are three and eleven times more active than cimetidine — the benchmark compound at the time of invention (Figure 2). After selecting these two lead compounds, Danbury continues, it would have been obvious to combine the polar tail from example 44 (Figure 3) with the substituted heterocycle from tiotidine (Figure 4), thus creating the intermediate compound (Figure 5). Thereafter, to create famotidine, Danbury argues that it would have been obvious to perform a bioisosteric substitution of the carbamoyl (CONH2) group in the intermediate compound with a sulfa-moyl group (S02NH2).
The district court correctly rejected Danbury’s argument. Specifically, Dan-bury did not show sufficient motivation for one of ordinary skill in the art at the time of invention to take any one of the following steps, let alone the entire complex combination: (1) selecting example 44 as a lead compound, (2) combining the polar tail
from example 44 with the substituted het-erocycle from tiotidine, and (3) substituting the carbamoyl (CONH2) group in the intermediate compound with a sulfamoyl group (S02NH2) to create famotidine.
At the outset, Danbury did not show the required motivation for selecting example 44 as a lead compound. Danbury’s assertion of motivation rests on the fact that example 44 is three times more active than eimetidine. That activity alone, however, is not sufficient motivation. As the trial court noted, other prior art references disclosed compounds with H2 antagonist activity up to ten times higher than cimeti-dine.
See Yamanouchi,
21 F.Supp.2d at 373. If activity alone was the sole motivation, other more active compounds would have been the obvious choices, not example 44.
Danbury also does not show the motivation to combine the polar tail of example 44 with the substituted heterocycle of tioti-dine, then to substitute the carbamoyl with a sulfamoyl. To show such motivation, Danbury argues only that an ordinary medicinal chemist would have reasonably expected the resulting compound to exhibit the baseline level of H2 antagonist activity. The baseline level of activity is a mere th the activity of eimetidine. This level of motivation does not show a “reasonable expectation of success.”
In re Longi,
759 F.2d at 897. The success of discovering famotidine was not discovering one of the tens of thousands of compounds that exhibit baseline H2 antagonist activity. Rather, the success was finding a compound that had high activity, few side effects, and lacked toxicity. As Danbury’s expert testified, the ordinary medicinal chemist would not have expected famoti-dine to have the “most desirable combination of pharmacological properties” that it possesses. J.A. at 1052.
Furthermore, the prior art offers no suggestion to pursue the particular order of manipulating parts of the compounds. Danbury’s proposed obvious course of invention requires a very specific series of steps. Any deviation in the order of combination would have taught away from famotidine. If, for instance, the sulfamoyl group were substituted for the carbamoyl group on example 44 without attaching the substituted heterocycle from tiotidine, the evidence showed that the resulting compound would have }iooth the activity of eimetidine. Famotidine, on the other hand, has 40 times the activity of cimeti-dine. Danbury offered no evidence suggesting what might have led an ordinary artisan in this field to follow the precise steps that produced a remarkable invention.
Danbury falls far short of satisfying its burden of showing a prima facie case for structural obviousness by clear and convincing evidence. Instead, as the district court aptly concluded, this case “has all the earmarks of somebody looking at this from hindsight.”
Yamanouchi,
21 F.Supp.2d at 370. Because Danbury did not show even a
prima facie
case for obviousness, this court has considered, but need not separately address, the strong objective evidence of non-obviousness.
B.
On appeal, Danbury contends that the district court improperly rejected its request to examine the inventor of the ’408 patent, Dr. Yanagisawa, and thereby abridged its right to be fully heard. At trial, Danbury proffered that Dr. Yanagisawa’s testimony would show an alleged distortion in his data provided to the U.S. Patent and Trademark Office. The alleged distortion, according to Danbury, would prove that tiotidine is actually more active than famotidine.
Upon review, this court concludes that the district court did not abuse its discretion in excluding Dr. Yanagisawa as a witness. Under Rule 52(c), “the right to be ‘fully heard’ does not amount to a right to introduce every shred of evidence that a party wishes, without regard to the probative value of that evidence.”
First Va. Banks, Inc. v. BP Exploration & Oil, Inc.,
206 F.3d 404, 407 (4th Cir.2000);
see Gran
ite State Ins. Co. v. Smart Modular Tech., Inc.,
76 F.3d 1023, 1031 (9th Cir.1996). Moreover, the advisory committee notes clarify that the rule “authorizes the court to enter judgment at any time that it can appropriately make a dispositive finding of fact on the evidence.” Fed.R.Civ.P. 52 advisory committee notes (1991 amendment).
In the present case, the district court made a dispositive finding that the record showed no motivation to combine the prior art in the way suggested by Danbury. Dr. Yanagisawa’s testimony would not bear on this dispositive finding. Rather, Dan-bury’s own expert, Dr. Loev, testified that famotidine is more active than tiotidine — a position directly contrary to what Danbury stated would be shown by Dr. Yanagisa-wa’s testimony. Further, Danbury, on appeal, admitted that it had an opportunity to notice and depose Dr. Yanagisawa, but did not do so.
Accordingly, this court agrees with the district court’s finding that Danbury was fully heard within the meaning of Rule 52(c). This court therefore affirms the district court’s grant of JMOL, sustaining the validity of claim 4 of the ’408 patent.
III.
This court reviews the district court’s decision to award attorney fees under an abuse of discretion standard.
See Avia Group Int’l., Inc. v. L.A. Gear Cal., Inc.,
853 F.2d 1557, 1567, 7 USPQ2d 1548, 1556 (Fed.Cir.1988). Under that standard, this court affirms the district court’s decision unless the court’s decision is clearly unreasonable, arbitrary or fanciful, or based on an erroneous conclusion of law or fact.
See Heat & Control, Inc. v. Hester Indus., Inc.,
785 F.2d 1017, 1022, 228 USPQ 926, 930 (Fed.Cir.1986) (citations omitted).
At the outset, this court recognizes that the Hatch-Waxman Act authorizes an award of attorney fees to the prevailing party in exceptional cases on the basis of an ANDA filing. As an initial matter, title 35 recognizes that the submission of an ANDA “shall be an act of infringement ... if the purpose of such submission is to obtain approval under such Act
to engage in
the commercial manufacture, use, or sale of a drug ... claimed in a patent ... before the expiration of such patent.” 35 U.S.C. § 271(e)(2) (emphasis added). Thus, under the terms of the Act, an ANDA filer may in fringe without even engaging in any actual commercial activities. The mere act of filing an ANDA constitutes infringement.
The Act also permits an award of attorney fees for infringement by an ANDA filing. Section 271(e)(4) states: “For an act of infringement described in paragraph (2) ... a court may award attorney fees under section 285.” Section 285, in turn, provides: “The court in exceptional cases may award reasonable attorney fees to the prevailing party.” 35 U.S.C. § 285. The “paragraph (2)” infringement specified in § 271(e)(4) is the filing of an ANDA. Accordingly, the Act unambiguously permits an award of attorney fees to the prevailing party in exceptional cases on the basis of an ANDA filing.
In fact, the Act singles out paragraph (2) infringement - ANDA filings - as a basis for an attorney fee award. In the same section, the Act allows damages and other monetary relief “only if there has been commercial manufacture, use, offer to sell, or sale within the United States ... of an approved drug.” 35 U.S.C. § 271(e)(4)(C). Yet with regard to paragraph (2) infringement, the Act incorporates no such restriction, but rather authorizes fee awards. Accordingly, the Act itself does not limit an award of attorney fees for paragraph (2) infringement to cases involving infringing commercial sales.
See
35 U.S.C. § 271(e)(4).
As
noted above, section 271(e)(4) authorizes fee awards for paragraph (2) infringement in accordance with the standards for section 285 exceptional cases. This court, in turn, has recognized many varieties of misconduct that make a case exceptional for a fee award. These forms
of misconduct include willful infringement, see,
e.g., Avia,
853 F.2d at 1567;
Rosemount, Inc. v. Beckman Instruments, Inc.,
727 F.2d 1540, 1548, 221 USPQ 1, 8-9 (Fed.Cir.1984), inequitable conduct before the PTO, offensive litigation tactics, vexatious or unjustified litigation, or frivolous filings,
see Hoffmann-La Roche Inc. v. Invamed Inc.,
213 F.3d 1359, 1365, 54 USPQ2d 1846, 1850 (Fed.Cir.2000);
Beckman Instruments, Inc. v. LKB Produkter AB,
892 F.2d 1547, 1551, 13 USPQ2d 1301, 1304 (Fed.Cir.1989).
In the present case, the district court determined that Danbury’s conduct amounted to willful infringement.
See Yamanouchi,
21 F.Supp.2d at 376. An ANDA filing by its very nature is a “highly artificial act of infringement,” therefore, the trial court need not have elevated the ANDA certification into a finding of willful infringement.
See Eli Lilly & Co. v. Medtronic, Inc.,
496 U.S. 661, 678, 110 S.Ct. 2683, 110 L.Ed.2d 605, 15 USPQ2d 1121, 1130 (1990). Rather, Danbury’s misconduct in filing a wholly unjustified ANDA certification and misconduct during the litigation that followed warranted the district court’s finding that this case was exceptional.
The joint operation of §§ 271(e) and 285 require the paragraph (2) infringer to display care and regard for the strict standards of the Hatch-Waxman Act when challenging patent validity. As already noted, the Hatch-Waxman Act authorizes challenges to the validity of patents in accordance with strict statutory requirements. Specifically, a paragraph IV filing requires “a certification, in the opinion of the applicant
and to the best of his knowledge,
[that] each patent ... for which the applicant is seeking approval ... is invalid.” 21 U.S.C. § 355(j)(2)(A)(vii)(IV) (emphasis added). The Hatch-Waxman Act thus imposes a duty of care on an ANDA certifier. Thus, a case initiated by a paragraph (2) filing, like any other form of infringement litigation, may become exceptional if the ANDA filer makes baseless certifications.
This court concludes that the district court’s finding that Danbury made a baseless certification is not clearly erroneous. In the first place, Danbury’s case for obviousness presented at trial contained glaring weaknesses, precipitating a JMOL. The ANDA certification notice and its supporting affidavits, upon which Danbury relies to show that it had a good faith belief in invalidity, suffer similar weaknesses. The certification statute requires notice to the patentee of “the factual and legal basis” of invalidity.
See
21 U.S.C. § 355(j)(2)(B)(ii). Danbury’s notice does not present a
prima facie
case of invalidity, and makes no reference to famotidine’s potency, safety, and lack of side effects, among other distinguishing properties accompanying its unusually high activity.
See Yamanouchi,
21 F.Supp.2d at 376. “Moreover, Dr. Loev admitted at trial that, as of 1992, he could not tell from [famotidine’s] chemical structure whether it would be toxic nor predict its lack of side effects. He further testified that he could not predict the effects on potency that would be caused by the structural manipulations he claimed to be obvious.”
Id.
When Danbury proceeded in the face of these weaknesses, its certification amounted to baseless and unjustified misconduct. In certifying invalidity, Danbury disregarded its duty to exercise due care.
In assessing whether a case qualifies as exceptional, the district court must look at the totality of the circumstances.
See Kaufman Co., Inc. v. Lantech, Inc.,
807 F.2d 970 at 978-79, 1 USPQ2d 1202, 1208 (Fed.Cir.1986). These circumstances include Danbury’s choice to produce during trial a 1993 opinion from its patent attorney, Mr. Alfred B. Engelberg. This legal opinion contained an acknowledged error in chemistry, which was critical to its conclusion of obviousness. Danbury’s own expert, Dr. Loev, conceded at trial that “En-gelberg’s interpretation of the [’408] patent was patently incorrect and that the [’408] patent nowhere described the formulation relied upon by Engelberg.”
See Yamanouchi,
21 F.Supp.2d at 376.
Based on the foregoing, the district court properly found that Danbury’s ANDA filing was “without adequate foundation and speculative at best.”
Yamanouchi,
21 F.Supp.2d at 376. The district court thus found the case “exceptional” and awarded attorney fees to Yamanouchi. This court detects no abuse of discretion by the district court in its award. This court therefore affirms the district court’s award of attorney fees to Yamanouchi.
CONCLUSION
Because the prior art does not render obvious claim 4 of the ’408 patent, this court affirms the district court’s grant of JMOL upholding its validity. Moreover, because the district court did not abuse its discretion in awarding attorney fees, this court affirms.
COSTS
Each party shall bear its own costs.
AFFIRMED.