In Re Bendamustine Consolidated Cases

Case Name: In Re Bendamustine Consolidated Cases, No. 13-2046-GMS, 2016 U.S. Dist. LEXIS 75624 (D. Del. June 10, 2016) (Sleet, J.)

Drug Product and Patent(s)-in-Suit: Treanda® (bendamustine HCl); U.S. Patents Nos. 8,436,190 (“the ’190 patent”), 8,609,863 (“the ’863 patent”), 8,791,270 (“the ’270 patent”), and 8,895,756 (“the ’756 patent”)

Nature of the Case and Issue(s) Presented: Cephalon markets Treanda, indicated for the treatment of patients with chronic lymphocytic leukemia and indolent B-cell non-Hodgkin’s lymphoma that has progressed during or within six months of treatment with rituximab or a rituximab-containing regimen. Cephalon accused seven parties of infringing the patents-in-suit. The limitations of the eleven asserted claims fall into three major categories: (i) concentrations and ratios of tertiary-butyl alcohol (“TBA”) and mannitol; (ii) vial size and reconstitution volume; and (iii) impurity levels. The issues raised in this opinion were whether: (i) the asserted claims were obvious; (ii) claims 1, 3, and 5 of the ‘270 patent were anticipated; (iii) claims 1 and 19 of the ‘270 patent were invalid due to an on-sale bar; and (iv) claims 19-21 of the ‘270 patent were invalid as derived under 35 U.S.C. § 102(f).

Why Cephalon prevailed: With regard to obviousness, the defendants argued that a POSA would start with Ribomustin, a prior-art product, and would have been motivated to reduce its degradants, reduce the required vial size, and improve its reconstitution (dissolving) time. They argued that to gain those benefits, a POSA would choose to create a lyophilized product, and would have selected a TBA/water co-solvent system for one step during lyophilization. Defendants argued that finding a useful lyophilization process and arriving at the claimed limitations would only require routine experimentation and that a POSA would have a reasonable expectation of success in so doing.

The court accepted defendants’ arguments that a POSA would have been motivated to improve Ribomustin’s vial size and improve its reconstitution time. But concerning impurity levels, the court said that the prior art did not disclose the amounts of degradants in Ribomustin or whether the impurities present in Ribomustin reached an unacceptable level. The court concluded that a POSA would need this information before going through the effort of reducing degradants. Accordingly, the court concluded that a POSA would not be motivated to reduce the level of degradants.

Turning to the choice of a TBA/water co-solvent system for lyophilization, most processes use water alone. The court found it “particularly telling” that one of the defendants’ experts knew of only one commercialized pharmaceutical product made using TBA as a co-solvent as of the invention date. The court also observed that no prior art employing TBA/water with bendamustine, or even with related compounds, was presented. The court considered a reference that used TBA/water with a nitrosourea compound. But because bendamustine is a nitrogen mustard compound, the court discounted that reference’s teachings. The court did credit a reference that presented a “multitude of problems” in the selection of solvent systems. As the court had already decided that reducing degradants was not a motivation, it discounted testimony about solvent selection based on that motivation. Ultimately, because lowering degradant levels was not a motivation, because water was the most common lyophilization solvent, and because of the known problems with the TBA/water co-solvent system, the court concluded that a POSA would not have been motivated to use that system in attempting to improve Ribomustin.

Concerning routine experimentation and a reasonable expectation of success, defendants argued that inventor needed only two and a half months to develop the claimed process. But the court said that this period was not indicative of the need for only routine experimentation because the inventor was actually in a position superior to a POSA. He had access to a wealth of confidential information on the formulation of Ribomustin, and therefore could make the informed choices necessary to obtain his final product with less experimentation than an outsider would require. Because the defendants failed to present a prima facie case of obviousness, the court did not consider the parties’ secondary considerations.

With regard to anticipation, the court found that the one purportedly-anticipating reference studied only relative impurity levels, not absolute levels. A second reference showed the claimed impurity levels, but it failed to meet other claim limitations. Thus, the court rejected the anticipation argument.

Next, the court found that the claimed product was not the subject of an on-sale bar. Because Ribomustin was never marketed in the U.S., the defendants offered a license agreement between Salmedix, the inventor’s employer, and Fujisawa, a manufacturer of Ribomustin, as evidence of a commercial sale. As part of the license, Fujisawa provided Salmedix with several batches of Ribomustin. The court examined the language of the license agreement and how the scientists at Salmedix used the Ribomustin batches they received, and concluded that the license was a development license and not a commercial sale.

With regard to the § 102(f) argument, the court found claims 19-21 of the ’270 patent to be derived under 35 U.S.C. § 102(f). The defendants argued that batches of Ribomustin delivered as part of the license had impurities within the claimed limits. The fact that a “Common Technical Specification” that was delivered as part of the license agreement specified impurity levels above the claimed level was not dispositive because all eight batches analyzed contained impurity levels well below the claimed levels. Accordingly, the court rejected the patentee’s argument that defendants failed to prove conception or enablement of these claims.