October 4, 2018
Motivation to combine and reasonable expectation of success are separate requirements that must be independently considered. By capitalizing on shortcomings in the petitions related to separately arguing these issues, Genentech was able to save two of three patents covering its blockbuster drug Herceptin®.
Herceptin®, the brand name for the monoclonal antibody drug trastuzumab, has been federally approved since 1998 primarily for treating breast cancer. Herceptin® has been extremely successful for Genentech, earning approximately $2.5 billion in the U.S. and $6.5 billion globally in 2016.
Conventionally, trastuzumab treatments are given weekly. Genentech obtained two patents—U.S. Patent Nos. 6,627,196 and 7,371,379—that extend the dosing regimen to every three weeks. The ’196 patent was challenged in IPR2017-00804 by Hospira and Samsung Bioepsis (via joinder with IPR2017-01958), and in IPR2017-01139 by Celltrion. The ’379 patent was challenged in IPR2017-00805 by Hospira and Samsung Bioepsis (via joinder with IPR2017-01959), and in IPR2017-01140 by Celltrion. The issues for the ’196 and ’379 patents overlap substantially and will be discussed together, grouped by challenger.
Hospira and Samsung Bioepsis Petitions
Hospira and Samsung Bioepsis challenged the patents based on the old Herceptin® label in combination with two secondary references and the knowledge of one skilled in the art. Petitioners argued trastuzumab was widely known as an efficacious breast cancer treatment, as established by the three references. Petitioners then argued that, although the references only teach a weekly regimen, a skilled artisan “would nonetheless have been motivated to decrease the frequency of trastuzumab administration to once every three weeks for several reasons.” This argument was based on general knowledge that a more convenient (i.e., less frequent) dosing regimen would improve patient compliance and quality of life for terminally ill patients. And, a tri-weekly regimen would align with chemotherapy dosing schedules, reducing the number of trips necessary to the clinic. Petitioners also argued that arriving at the tri-weekly dosing schedule was merely a matter of routine calculation and optimization, supported by sample pharmacokinetic calculations.
The Board summarized the two main issues in the proceedings as (a) whether there would have been a motivation to extend the weekly dosing interval taught in the prior art to a tri-weekly dosing interval based on concerns about patient convenience and quality of life; and (b) whether there would have been a reasonable expectation of success in implementing this regimen based on petitioners’ expert’s pharmacokinetic analysis. The Board emphasized that these are distinct legal requirements and addressed these issues separately, citing Intelligent Bio-Systems, Inc. v. Illumina Cambridge, Ltd., 821 F.3d 1359, 1367 (Fed. Cir. 2016) (“One must have a motivation to combine accompanied by a reasonable expectation of achieving what is claimed in the patent-at-issue” and “the Board conflated two different legal concepts—reasonable expectation of success and motivation to combine”).
The Board was persuaded with regard to motivation to combine, noting that a skilled artisan “would have been motivated to extend the dosing interval for the simple (yet compelling) reasons that doing so would have been more cost-effective and less burdensome for the patient.”
But, the Board found that Hospira and Samsung Bioepsis did not establish a reasonable expectation of success based on their proposed pharmacokinetic analysis allegedly showing that efficacy of trastuzumab could be maintained with a three week dosing schedule. Genentech persuasively argued against these calculations, emphasizing that the calculations assume linear kinetics, but trastuzumab may have non-linear kinetics, and thus the relevant pharmacokinetic parameters do not remain constant. Genentech also reiterated there was nothing in the prior art to allow one skilled in the art to accurately predict whether the three-week dosing would be effective.
The Board agreed, noting the relative novelty of using antibodies as cancer treatments as of the priority date and the lack of any prior art references discussing the feasibility or viability of a tri-weekly antibody dosing regimen. Moreover, while petitioners’ calculations were based on “textbook” equations that were known in the actual pharmacokinetic analysis was not based in the prior art and considered to be “largely based on impermissible hindsight.” Furthermore, the prior art established that trastuzumab exhibits dose-dependent kinetics, and the very textbook relied upon notes that “dose-dependent and time-dependent kinetic behaviors defy easy quantitative description and prediction.” Petitioners submitted substantial evidence in their Reply attempting to support their theory, but the Board was unpersuaded, particularly based on contradictory statements in the evidence regarding uncertainty and attempts to generalize small molecule kinetics with antibody kinetics. Accordingly, the Board concluded that Petitioners did not establish the reasonable expectation of success required for obviousness.
Lessons for Practitioners
This case illustrates the importance of arguing and attacking motivation to combine and reasonable expectation of success separately. For petitioners, reasonable expectation of success must be clearly articulated with no hindsight bias, and care should be taken when extrapolating models to cover alternative scenarios. Genentech’s patent owner response provides a good example of how to persuasively attack reasonable expectation of success arguments, particularly with regard to drugs.
The Celltrion petitions provide an example of where these issues were not sufficiently argued separately. Here, the Board found that Celltrion did not sufficiently establish a motivation to combine the references, particularly with regard to modifying the claimed loading and maintenance doses. And because the Board found that Celltrion had not met its burden to show that an ordinary artisan would have been motivated to modify the dosage amount in the first instance, its reasonable expectation arguments, premised upon efficacy associated with the modified dosage, also failed. Moreover, Celltrion attempted to establish reasonable expectation of success for claims 24, 25, 29 and 30 in a single paragraph referring back to its analysis of claim 1. But those claims recited a different dosage and thus the arguments—which failed anyway—were insufficient.
Hospira (IPR2017-00737), Samsung Bioepsis (IPR2017-01969 by joinder with Hospira), and Celltrion (IPR2017-01122) were successful in challenging Genentech’s U.S. Patent No. 7,892,549, a third patent covering Herceptin®.