Hospira UK Limited v Genentech Inc, Patents Court, 10 April 2014,  EWHC 1094 (Pat), Birss J
The Patents Court found two of Genentech’s patents for Herceptin (trastuzumab) invalid, and granted Hospira a declaration of non-infringement in respect of certain trastuzumab formulations. One patent was for a dosage regimen and the other for a purified composition of trastuzumab. Birss J’s judgment contains some significant legal points, particularly in relation to the dosage regimen patent.
The dosage regimen patent claimed a three weekly administration of trastuzumab, in contrast to the weekly dosage in use at the priority date. One interesting issue was on claim construction. Claim 1 was in Swiss-type form (“use of X in the manufacture of a medicament for the treatment of Y”). The parties agreed that: (i) in the context of a Swiss-type claim, “treatment” means something which is indeed an effective treatment – in this case for breast cancer; and (ii) in the context of a second medical use patent, “for” does not mean “suitable for” but rather “suitable for and intended for”.
Birss J noted that these aspects were necessary for a claim to be novel over a mere proposal to administer the drug in the manner claimed. He observed that clinical trial data was not generally required but plausibility was required by both the EPO (T609/02 Salk) and the UK courts (Regeneron  EWCA Civ 93).
Birss J held that patent was obvious over the prior US FDA label for Herceptin; because that described a weekly dosage regimen combined with administration of paclitaxel every three weeks, the skilled team would see no reason not to conduct a trial of a three weekly regimen.
Birss J held, obiter, that if the patent did involve an inventive step it would be bad for insufficiency, noting that, because the specification must contain enough information to make it plausible that the invention will work across the scope of the claim, on the facts of this case the key practical question was the same as for obviousness: would the skilled team conduct the trial? He also held that for priority the overall requirement was an “enabling disclosure” (Biogen  RPC 1), which included “plausibility” (reluctantly disagreeing with the TBA decision in Gemvax (T903/05) that plausibility was not required), and concluded that requirement was not met.
The second patent in issue claimed a composition comprising trastuzumab and certain impurities in an amount of less than “about 25%”. In common with recent decisions on numeric limitations, the parties agreed that meant less than 24.5%. Birss J held that all the claims of this patent were anticipated and/or obvious in light of several items of prior art. Genentech also has a compound patent to trastuzumab, which Hospira did not challenge and for which the UK SPC expires in July 2014. Both patents at issue in this case had been held invalid by the EPO Opposition Division but appeals to the TBA are pending. It remains to be seen whether Genentech will appeal this UK decision and if so, whether they will try and prevent Hospira launching their generic trastuzumab medicines pending that appeal
Read the judgment (in English) here.
Head note: Rowena Stent