On October 17, 2016, the PTAB denied institution of three IPRs [IPR2016-00912, IPR2016-00915, and IPR2016-00916] petitioned by Swiss Pharma AG against three patents owned by Biogen IDEC directed to its anti-α4 integrin antibody product, TYSABRI (natalizumab), marketed to treat multiple sclerosis and Crohn’s disease. The PTAB weighed the Petitioner’s assertions of routine experimentation against the Patent Owner’s arguments of unpredictability in deciding the outcome.

The three patents (US 8,815,236, 8,349,321, and 8,900,577) each claim priority to the same provisional application filed in 2003, and are directed to formulations containing either 20 mg/ml natalizumab (‘321 patent, claim 1) or 20 to 150 mg/ml natalizumab (‘577 patent, claim 1), and methods of using the latter natalizumab formulation (‘236 patent claim 1). The antibody formulations in the claims also include about 10 mM phosphate buffer, sodium chloride and polysorbate 80. Natalizumab was approved in 2004 for the treatment of multiple sclerosis and a previous formulation used in clinical trial was at a concentration of 5 mg/ml. The patents disclose that previous formulations could precipitate during dilution and dosing of the antibody.

Swiss Pharma petitioned for IPR on the grounds that the claims of the patents were obvious over two different combinations of prior art, arguing that the buffers in the formulation are known in the art and the amount of natalizumab recited in the claims is obvious in view of other compositions of antibodies disclosed in the cited art. In support, Petitioner cited a publication (Van Oosten) that discloses administration to MS patients of a formulation containing 10 mg/ml of a humanized mouse monoclonal anti-TNF antibody , and an article (Sorbera) that describes administration to MS and inflammatory bowel disease patients a formulation containing up to 3 mg/ml of the same antibody. The Petitioners also cited an excerpt from the Physician’s Desk Reference (PDR) for its disclosure of formulations of the antibody Daclizumab at a concentration of 5 mg/ml.

In a second combination of references, Petitioner cited to (i) a publication (Gordon) which discusses a trial in which natalizumab was used in a 5 mg/ml formulation in histidine buffer and polysorbate 80, (ii) another PDR reference to Orthocolone formulated at 1 mg/ml in buffer, and (iii) a study in monkeys (Aversano) in which an antibody against CD18 was administered in a 5 mg/ml buffered solution.

Petitioner argued that in view of the disclosures in the art showing that high concentrations of antibody compositions were known, achieving the claimed concentration of natalizumab was nothing more than routine optimization of a result effective variable. Petitioner also argued that choice of buffers in the claimed formulations was not inventive because one buffer, e.g., histidine buffer as disclosed in Gordon, could readily be substituted for another buffer, e.g., phosphate buffer, and asserted that simple substitution of one active ingredient (i.e., antibody) for another in a formulation was routine.

The Patent Owner (PO) argued that none of the cited references actually discloses natalizumab at a concentration of 20 mg/ml, and that it was well-known in the art at the time that it was difficult to achieve an antibody formulation that was stable at such a high concentration as antibody formulation could be unpredictable. PO further argued that Petitioner failed to show that it would have been routine to achieve a natalizumab formulation at such a high concentration. In particular, PO argued that the Petitioner’s identification in a publication of one example of an antibody at 50 mg/ml was formulated in saline for in vitro testing and the publication presented no evidence of the formulation’s stability. PO also offered evidence that antibody formulation was not routine or standard in the art and argued that Petitioner did not indicate what evidence in the art motivated one of skill to reach the claimed concentration.

The Board agreed with PO’s position, asserting that there was no evidence as to why one of skill would modify the concentration of natalizumab to be four times that previously disclosed in the art. The Board also stated that Petitioner failed to provide evidence of how routine optimization would lead to the recited concentration of natalizumab in the claims. The Board, however, offered no guidance on what evidence would have been “sufficient” to demonstrate that the adjustment of the antibody to a concentration of 20 mg/ml would have been routine. According to the Board, the Petitioner failed to demonstrate a likelihood of success that any claims are obvious over the cited art and, therefore, the Board denied all three petitions.

From the decision, it seems the Petitioner’s cited publications and expert declarations were not sufficiently on point to sway the Board that the claims were likely obvious, and Petitioner failed to connect the disclosures in the cited art to provide a means to achieve the claimed antibody concentration. Without knowing what would be “sufficient” evidence of routine optimization, both Petitioners and Patent Owners are still left wondering what types of evidence will sway the Board to decide whether something is routine or, in contrast, unpredictable.