The UK Supreme Court handed down the judgment in a much anticipated biotech decision on 24 June. The Court found that Regeneron’s patents are invalid for lack of sufficient disclosure.
The Supreme Court therefore disagreed with the earlier Court of Appeal’s decision which – contrary to the first instance judge of the Patents Court – upheld the validity of the patents.
At the heart of the case is the legal requirement for sufficient disclosure in a patent. According to requirements under the European Patent Convention (EPC) and the UK Patents Act, a patent must include enough information so that a skilled person can work the claimed invention across the breadth of the claims without undue burden.
This decision is an important one, especially for the biotech industry. The judgment provides a review of the relevant case law and the principles that should be applied when considering sufficiency.
A patent with a broad claim is vulnerable if the claim covers embodiments that could not have been worked at the relevant date of the invention. To avoid third party challenges, Patentees need to carefully ensure that it is at least reasonable likely that all products covered by a product claim can be made based on the disclosure in the patent.
The case is likely to have an impact on future cases where there is doubt whether the patent provides enough disclosure to support the claims.
What is the technical background?
The decision concerns the GB national parts of European Patent EP1360287 and its divisional EP2264163.
The patents relate to transgenic mice that produce hybrid antibodies containing human variable immunoglobulin regions and mouse immunoglobulin constant regions. Also covered are methods for making such animals. The transgenic mice are useful in the production of therapeutic antibodies. To make the animals, mouse gene sequences (the murine VDJ regions) need to be deleted from the chromosomal immunoglobulin locus in the mouse genome and replaced with human immunoglobulin gene sequences (human VDJ regions).
The end result is a hybrid mouse with a so-called “reverse chimeric chromosomal locus”. This includes human immunoglobulin gene sequences (human VDJ regions) and retains specific mouse gene sequences (the constant regions). In this way, any antibody product that is made by the modified chromosomal locus in the mouse genome has certain human and certain mouse parts. Whilst transgenic mice with different genetic modifications that were available at the priority date had immunological deficiencies, mice made using the “reverse chimeric locus” do not show these problems.
What is the legal background?
As mentioned above, the legal principle at stake here is that of sufficient disclosure. A patent is often described as a bargain between the Patentee and the public. The essence of that bargain is that the Patentee makes full disclosure of the invention in return for a time-limited monopoly over its use. As part of this full disclosure, the patent must contain sufficient information to enable a skilled person to make or carry out what is claimed.
This principle is enshrined in the EPC and in the UK Patents Act. A Patentee must be able to demonstrate that a skilled person can make a claimed product using the teaching disclosed in the patent coupled with the common general knowledge which is already available at the time of the priority date. In doing so, the skilled person must not undertake an undue experimental burden or apply any inventiveness of their own.
First instance Decision at the High Court
Regeneron brought claims against Kymab for infringement of the two patents. Kymab denied infringement and counter claimed for revocation.
It was held that the claims of the patents were invalid because they lacked sufficiency. On the law relating to sufficiency, the judge noted the established principle that if the invention does not work with substantially all of the products or methods falling within the scope of a particular claim, then that claim will be insufficient. The critical issue here related to the length of the sequences swapped in and out of the mouse genome to create a mouse chromosomal immunoglobulin locus that includes human and mouse immunoglobulin sequence.
The judge held that the claims covered the deletion of a large stretch of mouse sequences in the mouse genome as well as the insertion of large human sequences. A specific problem for Regeneron was that it was found that Example 3 of the patent, as described, could not have been performed by the skilled person. Expert evidence showed that swapping sequences that are of a large size were far beyond the capabilities of the skilled person at the priority date. Indeed, it was noted that at the upper limits of the claim, such replacements were perhaps even beyond current capabilities. At the priority date (2001), the skilled person would only have had the ability to successfully swap in and out much smaller sequences. Even allowing for the fact that this was a technical field where a great deal of trial and error was often called for, this fell far short of what would have been required to put the invention into practice across the claim’s breadth.
It was therefore held that the claims were insufficient as they did not enable the genetic modification across the breadth of the claim (i.e. all sizes of genetic sequence modification in the mouse genome) without undue burden and without invention.
On infringement, as is often the case, the issue came down to one of claim construction, specifically the term “in situ replacement” in the claims. The judge held that “in situ replacement” of gene segments includes the case where a sequence is deleted and also when it is moved to a different location and inactivated. On that basis, it was held that Kymab would infringe claims of both patents had they not found to be invalid.
Second instance Decision at the Court of Appeal
Regeneron appealed the decision for invalidity and Kymab cross-appealed the finding that the claims would be infringed if valid.
The Court overturned the first instance decision on validity, finding that the claims of Regeneron’s two patents were in fact valid. It also held- consistent with the first instance decision- that the claims were infringed by Kymab.
The invention in the patents
The judgment includes a lengthy discussion of what constitutes the invention in the patents and considers that the invention disclosed in the two patents in suit has two principal aspects. A first aspect was thought to be the in situ replacement of mouse variable region immunoglobulin gene segments with human variable immunoglobulin gene segments, maintaining certain part of the mouse genome (the mouse constant regions), so create the “reverse chimeric locus”. A second aspect of the disclosure concerned methods which can be used to target, via homologous recombination, and modify, endogenous genes and chromosomal loci in eukaryotic cells
Kymab argued that for the question of sufficiency, the range of sequences that could be swapped in and out of the mouse genome was critical because this affects the ability of the mouse to deliver a spectrum of useful antibodies. Regeneron’s view was that the range was irrelevant because the invention was in fact the “reverse chimeric locus”.
The Court of Appeal agreed with this view and held that technical contribution of the patents was the use of a “reverse chimeric locus”. This was considered by the Court a “principle of general application” and a ground breaking invention which provided a radical departure from the disclosure in the prior art.
In other words, the invention provided a blueprint or springboard to embodiments that could not have been made at the priority date but were nevertheless enabled by the invention. Simply put, the Court of Appeal thought it was unfair to limit Regeneron’s monopoly for the invention that was considered to be “radical” and a “principle of general application” to only those types of hybrid mice which could be made when the patent was filed.
The Court of Appeal specifically noted that if the claim was limited to only those embodiments enabled at the priority date, then protection would rapidly become ineffectual. Therefore, it upheld the patents over a range of mice even though Regeneron could only make mice over a small part of the range, at the least beneficial end of the range with the smallest amount of human genetic material.
That said, the Court also noted that the skilled person could nevertheless have adapted Example 3 relating to the removal of mouse sequence and insertion of human sequence without undue effort and thereby produced transgenic mice falling within the claims of the patents.
In line with the first instance judgment, Kymab was held to infringe.
The Supreme Court Ruling
After a UK Court of Appeal ruling, the parties have the right to request permission from the UK Court of Appeal to mount an appeal in front of the Supreme Court. However, in this case, Kymab was refused permission by the Court of Appeal to appeal and appealed directly to the Supreme Court. The Supreme Court agreed to hear the case and Kymab’s tenacity was rewarded with a decisive win over Regeneron.
In a majority judgment (4:1), the Supreme Court overturned the earlier ruling and found that the patents were invalid for lack of sufficient disclosure.
The judgment reviews the governing case law and sets out 8 principles which guide the assessment of sufficient disclosure. The judgment also criticises the approach taken by the Court of Appeal in applying the law relating to sufficient disclosure.
One of the main criticisms was that the judgment of the Court of Appeal focused on the invention rather than the claims. The judge states that “Patents are about products and processes not pure ideas”. In other words, it was not enough that the benefits the invention unlocked would in due course be realised over the whole range. What mattered was whether a skilled person could, at the priority date, practice the invention across the whole range claimed (subject to de minimis or irrelevant embodiments). The judgment found that claims to a monopoly over the whole range of mice, i.e. mice that include very small or large inserts, went far beyond the contribution which the product made to the prior art at the priority date. This was in particular the case because the more valuable mice, i.e. those that contain large inserts of human gene sequence, could not be made at the priority date. A patent should enable substantially all products within the scope of the claim to be made by the skilled person at the priority date to prevent a Patentee from obtain an unjust monopoly. The disclosure in Regeneron’s patents however enabled the skilled person only to make products over a very small part of the range, let alone the most commercially useful products.
Regeneron’s patent family at the EPO
The patents at issue are members of a large family of European patents. A number of other patents granted in this family have been opposed at the EPO by Kymab and other Opponents and are currently in Opposition or Appeal proceedings. It will be interesting to see if the EPO takes note of the decision by the UK Supreme Court- what seems certain is that the Opponents will try and use the decision to their advantage to invalidate other patents in this family at the EPO.
The implications of the judgment
The issue of sufficiency of disclosure in biotech cases is often contentious, particularly in prosecution and Opposition proceedings of cases at the EPO.
Broad claims are vulnerable to attack
The Decision highlights that broad claims are vulnerable to attack if the skilled person is unable to make substantially all embodiments falling with in the claim scope when applying common general knowledge at the priority date. This does not mean that a Patentee has to demonstrate by way of examples that each and every embodiment has been tried and tested. It is enough to demonstrate that it is reasonably likely that the whole range of the claim scope can be practiced. However, if such argumentation is used, and evidence to the contrary comes to light in litigation, then a broad claim is likely to be invalid.
The Supreme Court focused on the claim, not the invention. In fact, this is where the Court of Appeal was said to have got it wrong Thus, even a ground breaking invention that has a “principle of general application” does not warrant a broad claim if not all embodiment falling within the claim can be worked at the relevant date.
When should a biotech patent be filed?
The decision also highlights another issue that biotech companies often face, that is the question of when to file an application. There are often pressures on companies to file an application early, for example due to requests from investors in fund raising rounds or because a publication will be made. At that stage, the availability of data showing that what is claimed can be worked across the range of the claim is often limited. There is a risk that filing early will therefore lead to issues with sufficient disclosure either during prosecution or in a third party attack. However, not proceeding with an early filing may simply not be an option, especially for early stage biotech companies.
Some of this risk can be mitigated by carefully considering arguments that can be made to support the scope of the claims. Also, a careful planning of further experiments that could yield supporting data for inclusion in the application within the 12 months priority period can help to strengthen a broad claim.
If you have any questions on this judgment, contact your usual attorney at Appleyard Lees or Barbara Fleck.
