This case, about a patent on the active ingredient in a blockbuster high blood pressure medicine, illustrates the importance of delving into the technology to ensure that obviousness is examined from the perspective of one of ordinary skill in the art at the time, in this case, a medicinal chemist. The defendants argued the patent was invalid for obviousness based on what defendants called a mere “single atom” difference between the invention, and a compound in the prior art. The district court found no obviousness and the Federal Circuit affirmed, finding that an ordinary medicinal chemist at the time would not have plucked the prior art compound from obscurity as a starting point and that, even assuming the alleged starting point, this difference of one atom gave the relevant chemical groups completely opposite properties.

 

 

In Daiichi Sankyo Co., Ltd. v. Mylan Pharm. Inc., 670 F.Supp.2d 359, 367 (D.N.J.2009), the district court was asked to rule on the accused infringers’ challenge that the patent claim to the chemical compound olmesartan medoxomil was obvious.

 

Olmesartan medoxomil was not the first in its class of angiotensin receptor blockers (ARBs); E.I. du Pont de Nemours Company, Inc. (DuPont) had discovered a first generation ARB, losartan. Id. at 363. DuPont’s work leading to losartan revealed the drug was most effective with “lipophilic” chemical groups at a particular position on the molecule. Id. at 371. Lipophilicity means to have an affinity for a fatty environment, as opposed to “hydrophilicity,” which refers to attraction to water. DuPont continued to try to make better ARBs, coming out with a breakthrough second generation compound called DuP 532, and a number of backup compounds, including the six compounds in DuPont’s U.S. Patent No. 5,137,902 (the ’902 patent). Id. at 371-4. DuPont’s second generation work followed its teachings and increased lipophilicity at the 4-position. Id. Other pharmaceutical companies also worked on second generation compounds and also followed the DuPont teachings. Id.

 

The Daiichi Sankyo scientists invented a different kind of ARB with olmesartan medoxomil. Olmesartan medoxomil has a hydrophilic group (a hydroxyisopropyl) at the 4-position—the opposite of a lipophilic group, which DuPont’s work had taught would give the best activity. Id. at 363-4. In addition, olmesartan medoxomil has a medoxomil prodrug ester at the 5-position, unlike any of the DuPont work. Id. Based on the existing DuPont teachings, those in the field would have expected olmesartan medoxomil to be a failure—but surprisingly it had 100 times the potency of losartan, when orally administered. Id.

 

 

Attacking the patent solely on the grounds of obviousness, Mylan argued that a skilled artisan easily would have obtained the claimed compound by starting with one of the 6 examples of the ’902 patent compounds, the closest prior art cited in the patent prosecution for the objective indicia analysis. Id. at 375-6. Mylan argued that the difference in structure between the claimed invention and a prior art compound in the ’902 patent was only a difference of an oxygen atom at the 4-position of the imidazole ring, and the medoxomil moiety at the 5-position of the imidazole ring. Id. at 377. Indeed, Mylan said that the difference between the active metabolite of the claimed invention and Example 6 was just one atom, which it argued was enough alone to demonstrate obviousness. Id.

 

But, Mylan’s argument relied on starting with the invention and then selecting the closest prior art structure. An ordinary artisan at the time of invention would not have had the invention compound as a guide. Instead, as explained by testimony from both sides’ experts, an ordinary medicinal chemist would have looked to the compounds with the best data packages as the starting points. Id. at 375.

 

The district court found, based on the “substantial data” related to the second generation ARB compounds from DuPont and other companies, that Mylan failed to establish “by clear and convincing evidence that a person of ordinary skill would have selected the ’902 patent compounds out of the numerous other second generation ARBs disclosed in the prior art.” Id. at 376-7.

 

The court found that one of ordinary skill could have “selected a lead compound from a ‘broad selection of compounds,’” including any of the other second generation ARB compounds, and nothing pointed to the ’902 patent compounds as a place to start. Id. at 376-7. The district court particularly noted the testimony of the medicinal chemistry experts as to the criteria to select a starting point. Id. at 375.

 

On appeal, Mylan argued that the ’902 patent compounds were a per se starting point, because, after olmesartan medoxomil was invented, the ’902 patent compounds could be selected out of the extensive prior art as the structurally closest compounds. Daiichi Sankyo Co. Ltd. v. Matrix Labs., Ltd., 619 F.3d 1346 at 1354 (Fed. Cir. 2010).

 

The Federal Circuit rejected this argument, noting that the proper analysis “must avoid hindsight bias; it must look at the state of the art at the time the invention was made to find a motivation to select and then modify a lead compound to arrive at the claimed invention.”Id. (emphasis in original). The district court’s analysis, relying on the evidence of how a medicinal chemist at the time would have chosen a starting point, was affirmed by the Federal Circuit, which noted that “[p]otent and promising activity in the prior art trumps mere structural relationships.” Id.

 

Mylan also argued that “the prior art may point to more than a ‘single lead compound for further development efforts,’” and that the ’902 patent compounds should have been found to be among a group of possible lead compounds. Id. (quoting Altana Pharma AG v. Teva Pharm. USA, Inc., 566 F.3d 999, 1008 (Fed. Cir. 2009)). The Federal Circuit agreed that more than a single lead was possible, affirming the district court’s finding that there were a number of promising leads, but the ’902 patent compounds were not obviously among them. At the time medical chemists were many other compounds with properties that would have been considered superior to the ’902 compounds. Id.

 

 

The district court also addressed Mylan’s argument that a medicinal chemist of ordinary skill would have been motivated to modify the lipophilic alkyls found at the 4-position of the ‘902 patent compounds. The court found that even “[a]ssuming that the ’902 patent compounds would have been selected as lead compounds,” Mylan failed to prove that an ordinary artisan would have modified the ’902 patent compounds as necessary to obtain olmesartan medoxomil. Daiichi Sankyo, 670 F. Supp.2d at 378.

 

Indeed, the work in the ARB area taught away from a hydrophilic hydroxyl group at the 4-position of the imidazole ring, as the district court held. Id. at 369-75. But, to see this required a view through the lens of medicinal chemistry. Defendants argued that the art did not teach away in explicit text accessible to the lay person. See Daiichi Sankyo v. Mylan, Def. Post Tr. Br., D.I.133, No. 2:06-cv-03462 (D.N.J. May 6^th, 2009), at 7. But, using the perspective of one of ordinary skill in the art, including a medicinal chemist’s understanding of data in the art, the district court found that the prior art taught away from the use of a hydrophilic group as in Daiichi Sankyo’s invention. Daiichi Sankyo, 670 F. Supp.2d at 379-80.

 

In the district court’s analysis, it first looked at a technical statement in DuPont’s losartan patent which Mylan had relied on as prior art. The statement attributed greater biological potency to compounds having a lipophilic group at the 4-position to those with a hydrophilic group. Id. at 370, and did so in technical terms that would have been difficult for a lay person to understand. The district court found that “[b]ased on this statement alone, a person of ordinary skill in the art, upon reading this language, would have been discouraged from using a hydrophilic group at the 4-position.” Id.

 

Next, the court looked to the data in the prior art, as the testimony had indicated a medicinal chemist would. Based on this date the court held that the art clearly discouraged hydrophilic groups, which olmesartan medoxomil did, and preferred their opposite, such as lipophilic groups. Id. at 371-373. For this inquiry, the court looked at the importance of lipophilicity using the different types of analyses that the testimony demonstrated a medicinal chemist would have applied, including “subseries” and “regioisomer” analyses as well as the fact that vast majority of disclosed compounds were lipophilic at the 4-position, as were most of the most active compounds. Id.

 

The district court also looked to what those working in the field at that time actually did in their own research programs. Id. at 371, 374. DuPont—a leader in the area—followed its own teachings that a lipophilic substituent at the 4-position was best in its own second generation research in which the groups at the 4 position are lipophilic. Id. at 371. Further, the other second generation ARBs from Merck, Ciba-Geigy, Eisai, and Takeda also had lipophilic substituents at the 4-position—again following and reinforcing DuPont’s teachings away from the hydrophilic group of the invention. Id. at 374.

 

The Federal Circuit found no error in the district court’s ruling that, even accepting the ’902 patent compounds as leads, the prior art taught away from using the hydrophilic group found in olmesartan medoxomil. Daiichi Sankyo, 619 F.3d at 1354-5. And, the Federal Circuit found no error in the district court’s characterization of the invention as not structurally similar to the prior art compound. Daiichi Sankyo, 670 F. Supp.2d at 378. The Federal Circuit noted that the “single oxygen atom” difference emphasized by Mylan, “is of greater significance than it superficially appears, as it is the difference between functional groups.” Daiichi Sankyo, 619 F.3d at 1356, fn. 4.

 

 

Defendants filed a petition for certiori arguing that the Federal Circuit had applied a rigid “lead compound test” which dictated the result. Daiichi Sankyo responded, demonstrating that the analysis applied by the District Court and the Federal Circuit simply followed the thought processes of a medicinal chemist of ordinary skill in the art at the time. The petition for certiorari was denied.

 

 

This case reinforced the importance of analyzing obviousness from the perspective of a person of ordinary skill in the art. This approach is the backbone of the obviousness analysis from the historical precedent through the flexible, common sense approach more recently described by the Supreme Court in KSR.