The Medicines Co. v. Mylan Inc.

Generic drug manufacturer was liable for patent infringement as a matter of law when the level of acceptable impurities described in the manufacturer’s ANDA included the range of acceptable impurities claimed in the patent, and its invalidity defenses—anticipation, obviousness, non-enablement, and inequitable conduct—were rejected by the court.

Winter 2014

GENERICally Speaking: A Hatch Waxman Litigation Bulletin

Case Name: The Medicines Co. v. Mylan Inc., Civ. No. 1:11-cv-1285, 2014 U.S. Dist. LEXIS 152433 (N.D. Ill. Oct. 27, 2014) (St. Eve, J.) (Generic drug manufacturer was liable for patent infringement as a matter of law when the level of acceptable impurities described in the manufacturer’s ANDA included the range of acceptable impurities claimed in the patent, and its invalidity defenses—anticipation, obviousness, non-enablement, and inequitable conduct—were rejected by the court.)

Drug Product and Patent(s)-in-Suit: Angiomax® (bivalirudin); U.S. Pat. No. 7,582,727 (“the ’727 patent”)

Nature of the Case and Issue(s) Presented:  Bivalirudin is a reversible thrombin inhibitor which functions as an anticoagulant. Specifically, bivalirudin is used to prevent blood from clotting during catheterization procedures and must be formulated at a high level of purity. The ’727 patent describes a process for synthesizing bivalirudin with minimal impurities in the following way: bivalirudin is dissolved into mannitol; a pH-adjusting solution is added to the resulting mixture to raise the pH to a proper level; and the mannitol and pH-adjusting solution are removed to form a final, pure bivalirudin product. The ’727 patent claims a bivalirudin product with a 0.6% maximum Asp9impurity level. The dependent claims call for stricter Asp9 levels and limits on additional impurities.

Mylan’s ANDA product is bivalirudin with up to a 2.0% Asp9 impurity level. In response to TMC’s infringement allegations, Mylan argued that the ’727 patent was not infringed and was invalid as anticipated, obvious and not enabled. Mylan further argued that the patent was invalid on the basis of inequitable conduct. The court found in favor of TMC.

Why TMC Prevailed: Each asserted claim of the ’727 patent requires “pharmaceutical batches” of bivalirudin. The Court found that “pharmaceutical batches” could encompass a single batch of product, as long as that batch “is representative of all commercial batches” and the total impurities “represent levels for all batches made by the same process.” Mylan argued that TMC triggered the on-sale bar when it sold two of five “stability batches” of Angiomax® prior to the ’727 patent’s July 27, 2007 critical date.  The Court disagreed. Although the batches selected as “stability batches” by TMC had Asp9 impurities levels below 0.6%, many of the other batches produced at the same time, by the same process, had Asp9 levels much higher than 0.6%. Thus, the batches sold were not representative of all commercial batches of the product and TMC had not made a commercial offer of the claimed invention.

Next, the Court found that the prior art disclosed all elements of the claims except for the requirement of “pharmaceutical batches” and a maximum Asp9 impurity level of 0.6%. Mylan relied on internal TMC documents that discussed the problems of high levels of Asp9 in bivalirudin solutions, but failed to provide any evidence that these documents were publically disclosed. Absent such a showing, TMC’s internal documents did not qualify as prior art and could not be used in the Court’s obviousness analysis. Mylan also pointed to the same “stability batches” that formed the basis of its anticipation arguments, but the Court rejected that argument too for the same reasons as it had in its anticipation analysis.

Mylan also argued that the problem of Asp9 impurities in bivalirudin products was a predictable problem that was known in the art. Mylan claimed that a person skilled in the art (“POSA”) would be motivated to measure the Asp9 levels of bivalirudin products and minimize the amount of impurity present. The Court found Mylan’s argument attenuated at best. First, a POSA would need to produce a bivalirudin product following the same process of TMC to recreate the problematic levels Asp9 in the solution. Then, that POSA would need to identify the exposure to a pH-adjusting solution as the source of the problem, ignoring prior art that taught other factors could be the source of impurity. Finally, a POSA would have to conclude that changing the mixing conditions would result in a more pure solution. The Court concluded that such a sequence was unlikely to occur and was unsupported by the prior art by clear and convincing evidence.

Regarding the issue of non-enablement, Mylan argued that the patent specification teaches a compounding process resulting in a batch of bivalirudin with an Asp9 impurity level of 1.5%, well above the 0.6% threshold recited in the claims. The Court found this argument unavailing. During the relevant period, TMC contracted out its compounding process to a third-party manufacturer. TMC explained that the failed batch resulted because the manufacturer did not follow the correct compounding process, not because TMC could not reproduce the invention. Mylan did not provide any evidence casting doubt on this explanation. Thus, Mylan did not prove that it was the patented process, not operator error, that caused the out-of-range impurities.

The Court also rejected Mylan’s inequitable conduct argument. Mylan submitted four separate grounds for inequitable conduct: (i) failing to disclose the impurity level of the 1.5% batch produced by the third-party manufacturer; (ii) misrepresenting the number of early Angiomax (“original Angiomax”) batches that contained Asp9 levels of more than 0.6%; (iii) failing to fully disclose the process used to produce original Angiomax; and (iv) failing to disclose certain of the third-party manufacturer’s employees as co-inventors. First, the court did not believe that the inventor’s failure to disclose the 1.5% impurity batch rose to the level necessary for inequitable conduct. Both inventors testified that they believed the 1.5% batch was irrelevant, since they believed the batch was produced as a result of operator error. Mylan’s second inequitable conduct defense was based on a table in the specification disclosing the mean, standard deviation, and maximum amount of impurity in the batches of original Angiomax. Mylan claimed that the table was misleading because, when a normal distribution was applied to the data, it appeared that fewer batches were beneath the 0.6% impurity threshold than there actually were—60% of batches compared to the 87% of batches actually under the threshold. The Court rejected this argument because there was no indication that the examiner would apply a normal distribution to the table. Nor was there any indication that the examiner would have rejected the application had a normal distribution been applied. Mylan’s third inequitable conduct argument alleged that the inventors ought to have disclosed the fact that, during the compounding process for original Angiomax®, the mix rate was increased after the addition of the base solution and that the base solution was sometimes added by utilizing a nitrogen pressurized hose. The Court found that Mylan did not prove the materiality of either step, nor did the patent’s specification mislead the examiner. Finally, the Court rejected Mylan’s fourth inequitable conduct argument that the contract manufacturer’s employees met with one of the inventors and suggested methods of reducing the amount of Asp9 impurities in the bivalirudin solution. There was no evidence that it was the contract manufacturer’s employees who made the suggestions at the meeting. Further, the inventors testified that the cause of the high Asp9 levels was not identified until a later date.

Finally, the Court held that Mylan infringed TMC’s patent as a matter of law. Relying on Sunovion Pharms., Inc. v. Teva Pharms. USA, Inc., 731 F.3d 1271 (Fed. Cir. 2013), the court reasoned that because Mylan sought FDA approval for a product within the scope of the claims, Mylan infringed the ’727 patent. Mylan’s ANDA described a bivalirudin product with Asp9 impurities between 0.0% and 2.0%. TMC’s patent claimed the same product with an Asp9 impurity threshold of 0.6%. Mylan could have excluded the claimed range from its application, but chose not to. Further, Mylan’s ANDA test batch resulted in a bivalirudin solution with impurities within the claimed ranged. Thus, Mylan infringed all the asserted claims as a matter of law.



Oren D. Langer


Managing Partner, New York Office
Member of Executive Board

Ryan M. Schultz


Pronouns: he/him

Back to Top