Therapeutic use claims: assessing 'effective' treatment

Several significant patent cases were heard by the UK courts over the course of the last year and common themes are emerging which impact on the life sciences industry, in particular in the context of second medical use claims and dosage regimes.

One such theme, which runs throughout the judgments in Hospira v Novartis [2013] EWHC 516(Pat) and Eli Lilly v Janssen Alzheimer Immunotherapy [2013] EWHC 1737(Pat), is the question of what amounts to a disclosure of ‘effective’ treatment and the consequences which follow when assessing entitlement to priority and sufficiency.

While the standard itself may be familiar to many in this industry, the criterion to be applied in the context of patentability was hotly contested by the parties and explored in detail by Arnold J, who heard both cases at first instance.

Background

Novartis is the proprietor of two dosage regime patents claiming the use of zoledronic acid for the treatment of osteoporosis by intravenous administration, at a range of dosages and with dosing intervals of at least six months. Hospira and Generics t/a Mylan sought to revoke these patents to ‘clear the way’ for launch of their own generic products. Lilly was another ‘clearing the way’ case concerning Janssen’s second medical use patent, which claimed a class of antibodies which bind to the amyloid-β peptide, for use in preventing or treating Alzheimer’s (and related) disease.

Claim construction

It was common ground in both cases that the term “for” in the claims should be construed as meaning ‘suitable for’ (or, in the Hospira case, required the dosage regimes to be ‘effective’). Arnold J held this to mean that the product does in fact achieve the claimed therapeutic efficacy. While acknowledging the importance of context, he concluded that the primary criterion for determining efficacy in Janssen’s patent was success in a Phase II trial. Further, he agreed with Lilly that Phase III trials (if available) are the “best guide” since the skilled team recognises them as “the gold standard for determining efficacy”.

Entitlement to priority

This was not contested in Lilly but considered in detail in Hospira. Arnold J found that Novartis’ patents lacked priority for two reasons, the first being failure to disclose the particular combination of features ultimately claimed in the patents. For example, there was nothing in the general disclosure of the relevant priority document to link the claims which covered a defined dose and dosing interval of zoledronic acid with the other integers, ie, treatment of osteoporosis and intravenous administration.

"While patentees may tend to favour early filing strategies, it is important to ensure that priority documents disclose the actual combination of features ultimately claimed in the patent."

Second, even though the priority document included Phase II studies, Arnold J decided there was no actual disclosure that zoledronic acid will be effective in reducing fractures in osteoporosis patients—the skilled team would appreciate that a Phase III trial was required.

Lack of priority was determinative of the outcome at first instance (it was common ground that the patents lacked novelty if priority was lost), a finding that was upheld by the Court of Appeal in Hospira v Novartis [2013] EWCA Civ 1663.

Sufficiency

In Hospira, Arnold J found that Novartis’ claims which were open-ended should not be interpreted as extending to any dose and any dosage interval but only to those which work. Nonetheless, since their potential scope was “quite broad” and placed an “undue burden on the skilled team to find out what doses and dosage intervals work”, he held the claims invalid for insufficiency as they were not enabled across their breadth.

Insufficiency issues were key in the Lilly case (since novelty and obviousness objections failed) and Arnold J set out a more detailed and structured approach. Citing the Court of Appeal decision in Regeneron v Genentech [2013] EWCA Civ 93 (which in turn applies well-established principles of European patent law) he confirmed that the patentee must demonstrate that the claimed therapeutic effect is “plausible”, adopting the following two-stage enquiry:

1. Determine whether the patent disclosure, in the light of the skilled team’s common general knowledge, makes it “plausible” that the invention will work across the scope of the claim; and

2. If satisfied, consider whether later evidence establishes that in fact the invention cannot be performed:

(a) Without undue burden at all (‘classical insufficiency’); and

(b) Across the scope of the claim without undue burden (‘excessive claim breadth/Biogen insufficiency’).

Arnold J concluded that the plausibility threshold in stage one was satisfied on the basis of in vivo data, but only in respect of antibodies which bind to the N-terminal portion of amyloid-β rather than any antibody, as claimed. So the patent was insufficient for excessive claim breadth, as in Hospira.

Arnold J then considered post-published evidence in stage two, concluding that Janssen’s patent was also classically insufficient given the failure of Janssen’s Phase III trials of its own antibody to amyloid-β. In doing so he acknowledged again that Phase III trials are a “better guide” to efficacy than Phase II.

Conclusion

While patentees may tend to favour early filing strategies, it is important to ensure that priority documents disclose the actual combination of features ultimately claimed in the patent. Interestingly, had this been disclosed in Novartis’ priority document, the Court of Appeal indicated (obiter) that “I think one would conclude that the patentee was teaching that the regimen would be effective”.

This suggestion that no more data was required may be of some comfort to patent applicants, who now face a potential dilemma between current moves towards greater transparency of clinical trials and the risk of destroying novelty in their own inventions.

Broadly drafted patent claims continue to cause difficulties for patentees in the wake of excessive claim breadth objections, particularly in the medical use field. Therapeutic use patents invariably require disclosure of appropriate experiments to survive allegations of insufficiency. Exactly how much and what type of data is required will be a question of degree based on the nature of the invention and the state of the art in any given case, and expert evidence will no doubt be key.

Nonetheless, while finding in Janssen’s favour on plausibility in the Lilly case, Arnold J’s suggestions that Phase II (ideally Phase III) clinical trials are required to establish efficacy seem concerning for patentees. This can be contrasted with the earlier appellate guidance in Regeneron that it is not always necessary for a patentee to provide clinical trial data or animal testing, and that patents are not insufficient merely because they do not demonstrate therapeutic efficacy if, nevertheless, there is enough information upon which to found a reasonable prediction.

Although granted permission to appeal Arnold J’s decision, it seems that Janssen’s appeal is not proceeding, so we will have to wait for further guidance from the Court of Appeal on the threshold for patentability of therapeutic use claims.

Jennifer Antcliff is an associate at Powell Gilbert. She can be contacted at: jennifer.antcliff@powellgilbert.com

Dennis Waller is an associate at Powell Gilbert. He can be contacted at: dennis.waller@powellgilbert.com