Hospira (UK) Limited v Genentech, Inc., England and Wales Court of Appeal (Elias, Kitchin and Floyd), London, UK, 6 February 2015, Neutral Citation Number: [2015] EWCA Civ 57

In this judgment the Court of Appeal in England and Wales decided a case regarding Genentech’s patent EP (UK) 1 210 115 relating to a dosing regimen for the anti-cancer drug trastuzumab (Herceptin). The Court of Appeal upheld the conclusion of the Patents Court (Birss J) that the patent was invalid for obviousness. At first instance Birss J also held that the patent was invalid for insufficiency and other grounds of invalidity based on loss of priority date, however, the Court of Appeal did not consider sufficiency or priority on appeal.

=================================

Genentech’s patent claimed a specific dosing regimen for Herceptin, which at the priority date was a known drug for treating breast cancer and already approved by the FDA. The specific dosage regimen claimed by the patent was an initial dose followed by at least two further doses at three weekly intervals. This differed from the FDA approved regimen (described in the ‘FDA label’ document) which was a lower initial dose followed by lower doses on a weekly basis.

The difference between the FDA label and the inventive concept was the three weekly dosing regimen and the fact this regimen was effective to treat breast cancer (the different initial dose was not key as it would have been safe and obvious to try).

The question therefore was whether the regimen claimed in the patent was obvious in light of the FDA label information. The obviousness case advanced was that the claimed dosing regimen was obvious to try, where it must be shown that the skilled team would embark on the necessary work with a “fair expectation of success” (Conor v Angiotech applied).

The Court of Appeal held that Birss J was entitled to find that the idea of three weekly dosing was something that would have occurred to the skilled clinician who, upon consulting the skilled pharmacokineticist, would have been advised that it was worth carrying out a clinical trial (with sufficiently good prospects of the three week regimen being an effective cancer treatment).

The skilled team was essentially being asked to carry out a ‘no/go’ decision on whether to carry out a clinical trial, with Birss J concluding that the outcome was a ‘go’ based on the evidence. Hospira relied on two sets of calculations from its expert which provided estimates based on the FDA label information. The first set of calculations established that Herceptin was likely to be effective as a breast cancer treatment if administered at three weekly intervals.

However, this was based on a three weekly regimen of a higher dose of the drug than claimed in the patent. Hospira also sought to rely on evidence that it argued showed it would be obvious to try a lower dose of the drug on the three weekly regimen.

Whilst Birss J was not convinced that the reasoning supporting this later evidence was ‘robust’, he considered this did not matter and it was wholly obvious to try lower doses even if this second set of calculations were rejected. The Court of Appeal rejected Genentech’s argument that the judge was not entitled to find the invention obvious without accepting both aspects of Hospira’s evidence.

The Court of Appeal highlighted that, irrespective of the second set of calculations, the field of clinical pharmacokinetics depended on the judgment or experience of the skilled pharmacokineticist and would not be based purely on calculations. Further, the fact that the first set of calculations and analysis conducted by Hospira’s expert were done on a cautious basis, including with a safety factor, supported a finding of obviousness based solely on the first set of calculations.

Ultimately, the Court of Appeal found that Birss J had ample material before him to conclude that the patent was obvious; based on the evidence the skilled team would have sufficient confidence to try the lower doses in a regimen which fell within the claimed invention. The appeal was dismissed. The patent was held to be invalid.

Read the judgment (in English) here.

Head note: James Agnew and Scott Parker, Simmons & Simmons