Hospira UK Ltd v Genentech, Inc.

Obviousness is a single statutory question assessed by weighing all relevant circumstances. Obvious to try is not a substitute test, but one consideration within that assessment, requiring a reasonable or fair prospect of success. The inquiry is flexible and does not impose a special, higher threshold for therapeutic claims. A claim that includes attainment of a therapeutic effect may still be obvious to try if that prospect exists. A decision to undertake research because it is justified does not, without more, prove a fair expectation of success. The appellate court should respect the trial judge’s multifactorial evaluation unless there is an error of principle.

Genentech appealed against Arnold J’s order revoking a European patent concerning the use of trastuzumab with a taxane to treat HER2-positive breast cancer. The judge found the claims novel but lacking inventive step, holding that the prior publication Baselga 97 gave the skilled person a fair expectation of success in conducting the proposed phase III trial.

The appeal challenged the legal standard for fair expectation of success, the relevance of the technical and clinical burden of undertaking the work, and the judge’s assessment from the skilled person’s perspective. The central issue was whether the judge had applied the correct flexible approach to obviousness.

Appeal dismissed. The Court of Appeal, in the judgment of Floyd LJ agreed with by Kitchin and Longmore LJJ, held as follows.

1. Obviousness involves one statutory question: whether the invention was obvious at the priority date. Obvious to try is not a substitute test. It is one consideration in a multifactorial assessment and must be supported by a reasonable or fair prospect of success. The appellate court must respect the trial judge’s evaluation unless there is an error of principle, as explained in Biogen Inc v Medeva Plc [1997] RPC 1.
2. The fair-expectation inquiry is flexible and fact-sensitive. There is no universal requirement that success be more or less self-evident, and no special rule applies merely because the claim includes a therapeutic effect. Such a claim may nevertheless be obvious to try if the necessary fair prospect of success exists. The reasoning in Conor Medsystems v Angiotech Pharmaceuticals [2008] UKHL 49 did not require a stricter approach.
3. A research programme may be regarded as justified because its potential rewards are substantial, even where the expectation of success is weak. That justification alone cannot establish obviousness. The court must assess the strength of the scientific reasoning predicting success.
4. The judge was entitled to find a fair expectation of success. Combination therapy was common; the drugs acted on different targets and had different modes of action; their toxicity profiles did not significantly overlap; and the evidence supplied an unusually strong scientific rationale. The prior results were sufficiently encouraging, and there were no convincing obstacles to the proposed trial.
5. The work involved significant time and expense, but it was not technically burdensome. The patent specification could be used to assess the nature and burden of the work, which remained within the skilled person’s technical capacity. The later excitement generated by the successful results did not show that the earlier skilled reader lacked a fair expectation of success. Arnold J had made no error of principle.

1. Court of Appeal (Civil Division) dismissed Genentech’s appeal. Kitchin and Longmore LJJ agreed with Floyd LJ.
2. High Court of Justice, Chancery Division, Patents Court, Arnold J, ordered revocation of the patent in [2015] EWHC 1796 (Pat), finding the claims novel but lacking inventive step.