The Court of Appeal revisits obviousness as Hospira succeeds in clearing the way for generic trastuzumab

Background

In a recent decision (Hospira v Genentech [2016] EWCA Civ 1185), the Court of Appeal has upheld the first instance decision of Arnold J (Hospira v Genentech [2015] EWHC 1796 (Pat)) finding the Swiss-form claim of Genentech’s patent relating to the use of the monoclonal antibody trastuzumab in combination with a taxane for the treatment of HER2-positive breast cancer, to be invalid for obviousness.

This case is the third, and seemingly final, part of a series of cases brought by Hospira against several Genentech patents protecting its monoclonal antibody product Herceptin for HER-2 positive breast cancer (also known as trastuzumab). A brief history of the long-running dispute between the parties can be found in our previous elexica article, discussing the second case where the Court of Appeal had upheld the first instance decision finding two formulation patents invalid for obviousness (Hospira v Genentech [2016] EWCA Civ 780). As a result of this latest decision, and pending any appeals to the Supreme Court, Hospira appears to have now fully cleared the way for a generic Herceptin product launch.

The patent in suit summarises the results of a clinical trial of trastuzumab in combination with a taxane and has Swiss-form claims directed at the use of that combination to treat HER-2 positive breast cancer. Taxanes are a class of chemotherapeutic agents which include paclitaxel and docetaxel. Hospira’s attacks on the validity of the patent were based on a single piece of prior art, a review article called ‘Baselga’ that described a phase III clinical trial of trastuzumab in combination with various chemotherapeutic agents, including paclitaxel. The paper did not disclose any results from the phase III trial, but did contain results from earlier phase I and II studies.

At first instance Arnold J made several observations on claim construction, and following Regeneron v Genentech [2013] EWCA Civ 93, held that requirements for "clinical benefit" and "clinical efficacy" in claim 1 were functional features of the claim, and so the actual attainment of the clinical benefit was a requirement of the claimed invention. Furthermore, it was common ground following Warner-Lambert v Actavis Group [2015] EWCA Civ 556 that the reference to "for" in the claim meant that it must be known to, or reasonably foreseeable by, the manufacturer that trastuzumab would be intentionally administered in combination with a taxane for the (clinically beneficial) treatment of HER-2 positive breast cancer.

At first instance, Arnold upheld the novelty of the patent over Baselga as the required mental element, ie the intention to administer the combination to provide the specifically claimed clinical benefit, was missing. In other words, Baselga did not anticipate the invention as clinicians could not intend to administer the combination to achieve the clinical benefit unless they knew that the clinical benefit would be achieved. The first instance judge’s finding on novelty was not in issue on this appeal. Genentech focussed its appeal on only obviousness.

Obviousness

At first instance on the question of obviousness, Arnold J held that the patent lacked inventive step and observed that even though the results of the phase III trial were not discussed in Baselga, the skilled person would have had a reasonable expectation of success in the phase III trial given the positive and encouraging results of the phase II trials and the xenograft studies described.

On appeal, Counsel for Genentech contended that the first instance judge had made two errors in his analysis, both of which were rejected. We summarise these below.

1. The wrong approach was taken as to what amounted to a fair expectation of success

It was common ground that it would not be obvious to the skilled person from reading Baselga that the combination would definately result in increased efficacy. Rather, the issue was whether the skilled person would have a fair expectation of successfully achieving increased efficacy if he were to conduct the clinical trial using the combination. On appeal, Genentech argued that Arnold J had considered only whether the existing results "justified" Genentech’s decision to proceed with the phase III trial. However, Genentech argued that following the approach in Saint Gobain v Fusion-Provida [2005] EWCA Civ 177, the question to be asked should be whether it was “more or less self-evident that what is being tested ought to work”.

Genentech developed this point further with a reference to Conor Medsystems v Angiotech Pharmaceuticals [2008] UKHL 49, [2008] RPC 28, a case involving an assessment of whether it was obvious to use a taxol-coated stent to prevent restenosis. In that case, Lord Hoffmann in the House of Lords had stated that the question to be decided was whether it was obvious that a taxol-coated stent would prevent restenosis and not whether it might have this effect. However, Floyd LJ agreed with Counsel for Hospira that too rigid an application of such a principle would leave no room for treating an invention as obvious when it was obvious to try. Furthermore, Floyd LJ held that imposing a requirement for there to be a very high expectation of success for claims which include as part of their technical subject matter a therapeutic benefit or effect, would amount to the creation of a lex specialis for such claims, where such an interpretation and application of the law would depart from and override the general principle. Instead, Floyd J emphasised how the test is a flexible one, based on asking whether there is, in all the relevant circumstances, a fair expectation of success.

Floyd LJ then went on to consider Genentech’s motivation for conducting the trial. It submitted that although a company may regard conducting a clinical trial to be justified, this does not necessarily mean that it conducted the trial with the required reasonable or fair expectation of success; companies often conduct clinical research where there is little or perhaps no expectation of useful result, because of the potential for very substantial rewards should the research, against all the odds, result in a successful treatment. Floyd LJ agreed with this analysis in principle, and stated that it would be wrong to treat an invention as obvious simply on the basis that a company would consider the trial to be justified. Genentech then submitted that the fact that a skilled person would carry out the trial proposed by Balsega did not mean that there was the necessary expectation of success, in contrast to what the first instance judge had found. Floyd LJ did not accept this submission. He considered that there was a sound basis for finding that there was a fair expectation of success in the data provided by Baselga; the fact that combination therapy was common; what was known about these particular drugs individually, together with the expert evidence given at trial.

2. The judge had failed to take proper account of the position of the skilled person

Genentech contended that in contemplating the teaching of Baselga at the priority date, the skilled man would have neither trastuzumab nor another anti-HER2 antibody to hand, and would have to balance the arduous task of formulating and testing the antibody with the uncertainties surrounding a successful outcome of the trial described in Baselga. Genentech had argued that the skilled person would have considered “the stakes too high and the odds too long”. Genentech further argued that the trial judge was wrong in stating that the patent itself assumed that the antibody formulation and testing were steps which a skilled person was capable of undertaking, and that the patent would be invalid for insufficiency if these steps imposed an undue burden on the skilled person. Genentech claimed that the judge had not properly considered whether the skilled person, without knowledge of the patent, would have proceeded with the burdensome work when all he had to hand was the level of expectation provided by Baselga.

Floyd LJ rejected this argument, stating that the trial judge had accepted that there was significant burden in reaching the point of being able to conduct phase III trials, but that did not mean that the trials described in Baselga were technically burdensome as they were within the technical capacity of the skilled person. He also agreed with Hospira that this is no more than a skilled person would have to do to accomplish the teaching of the patent. He held that the first instance judge had considered the nature and burden of the work when making his assessment of whether there was a fair expectation of success.

Comment

With his second leading judgment considering obviousness in the same series of cases, Floyd LJ has clearly indicated that the UK Courts will continue to resist the application of any "one size fits all" test for obviousness. Rather, the assessment should be multi-factorial in nature and take account of the particular facts of each case. Indeed, the only approach which is common in nearly every assessment of obviousness, is the flexible and stepwise Windsufing/Pozzoli test, as Arnold J followed at first instance in this case.

Historically, therapeutic use claims have been proven reasonably resistant to novelty attacks, as in this case (and not appealed). However, this case illustrates the UK Court’s reluctance to formulate a test under obviousness which might make a special case of such claims. As Floyd LJ commented in this case, “such claims may nevertheless be obvious if it is obvious to try to achieve that technical effect by making something within the claims, provided that there is the necessary fair prospect of success”.

With the conclusion of these proceedings the path to launch generic Herceptin products in the UK now seems clear. However, it will be interesting to see whether the same can be said of the law of obviousness and whether Genentech will contend that the UK approach requires the verdict of the Supreme Court.