Hospira v Genentech - Herceptin formulation obvious
16 December 2016
In July this year, the Court of Appeal handed down a decision in Hospira v Genentech  EWCA Civ 780 which upheld the decision by Birss J in November 2014 to revoke the two Herceptin formulation patents for obviousness and added matter.
Herceptin and trastuzumab
Herceptin is an important breast cancer drug and contains a monoclonal antibody active ingredient known as trastuzumab. At the priority date there were two general approaches to formulating antibodies such as trastuzumab: one was to produce a ready-to-use liquid formulation and the other was to produce a lyophilized (freeze-dried) formulation which can be made up into a sterile solution for use.
The two patents filed by Genentech were EP 1 516 628B and EP 2 275 119B. Both concerned lyophilised formulations of trastuzumab and were essentially based on choosing certain excipients for the formulation. The excipients chosen were trehalose as the lyoprotectant, histidine as the buffer, and polysorbate 20 as the surfactant.
Decision at first instance
At first instance Birss J found that both patents were obvious over a pair of documents (Carter) which disclosed that trastuzumab was in phase II clinical trials for breast cancer as a liquid formulation. Birss J held that a skilled team of a clinician and a formulator reading Carter in light of their common general knowledge (CGK) would have been motivated to produce a lyophilized version of trastuzumab, and the claimed formulation was simply the result of a necessary and routine screening programme to find a satisfactory combination of excipients. Histidine was an obvious buffer candidate for the pH range at which trastuzumab was most stable, and polysorbate 20 was an obvious surfactant candidate. For the lyoprotectant, Birss J held that trehalose would have been on a list of possible candidates and any concerns about toxicity and regulatory approval did not make testing trehalose inventive.
Hospira v Genentech on appeal
On appeal Genentech's main argument was that Birss J had erred in principle to accept Hospira's case that the claimed invention was obvious because it could be reached by the application of routine approaches using CGK. Genentech argued that formulating proteins was difficult and unpredictable, that there were no pointers in the CGK or prior art to the claimed formulation and that the formulation had a beneficial effect in terms of stability. There was nothing to provide the skilled person with the necessary fair expectation of success.
Genentech also submitted that the judge had misapplied the distinction between "would" and "could" and with reference to EPO jurisprudence, argued that the possible inclusion of something in a programme for testing in order to see if it works but without any expectation of success did not establish obviousness.
In the lead judgement by Floyd LJ, however, the appeal was dismissed.
Floyd LJ agreed that the skilled team knew that lyophilised formulations of proteins had been successfully made before, and that the three types of excipient (buffer, surfactant and lyoprotectant) were all part of the formulator's common general knowledge and were being used for their known purposes. There was nothing inventive in the screening approach and no evidence that trastuzumab presented any special formulating problems.
In dismissing the appeal Floyd LJ also considered the "could-would" argument put forward by Genentech. Floyd LJ noted that there was no need to establish "in every case that the skilled person would necessarily have arrived at the precise combination claimed. The skilled person may be faced with a range of obvious possibilities, making it statistically unlikely that he will settle on any one of them. They will all be obvious".
In this case, the screening methods were part of the CGK, the tests involved were routine, the excipients were CGK and there was no a priori reason why a successful lyophilised formulation could not be made. Floyd LJ held that the team may have had a reasonable degree of confidence that the screening methods would produce a formulation that will work but to require them to be able to predict in advance which would be the successful combination is wholly unrealistic. There was no invention in embarking on a screening process to pick out the "good from the bad".
Case details at a glance
- Decision level: Court of Appeal (Civil Division)
- Jurisdiction: England and Wales
- Parties: Hospira UK Limited (claimant/respondent) and Genentech, Inc (Defendant/Appellant)
- Citation:  EWCA Civ 780
- Date: 12 & 13 July 2016
- Full decision: http://dycip.com/2016ewcaciv780