In this update we look at recent developments in the patent scene and comment on what they mean for those operating in the technology sector, whether they are drafting or evaluating their own patents, or trying to steer around those belonging to competitors.
Claim scope – what does the patent really cover?
Many fields of technology are heavily patented. But how do you work out whether a patent covers what you plan to do? Several recent UK court decisions have looked at the question and provide useful guidance.
ASSAI sued UK telecoms and broadband provider, BT, for infringement by its Next Generation Access (NGA) system of two patents for ADSL technology. ASSAI and BT agreed on the correct approach to claim construction, derived from the earlier cases (Kirin Amgen v Hoechst Marion Roussel, (2004) and Virgin Atlantic Airways v Premium Aircraft Interiors (2009), but not on how this should apply to these patents.
One patent dealt with the management of line profiles in a DSL system. The parties disagreed on the meaning of several claim features, BT relying on examples and detail from the text of the patent to narrow the meaning of the claims. The Court of Appeal said that this approach was wrong – "the patentee may have deliberately chosen to claim more broadly than the specific examples, as he is fully entitled to do". BT’s system infringed.
A second patent claimed a method for "generating" margin-related parameters for a modem pair. BT argued that selecting a margin related parameter from a table of values did not come within the claim requirement of "generating". They said that it would be illogical for a wider claim construction to be used if it meant that the claim would be obvious over the prior art. But the purpose of the invention as explained in the patent did not support this analysis. The reader would not think that use of the word ‘generating’ was intended to exclude use of a lookup table to choose a predetermined value.
Claim 1 of this patent was not valid, but a narrower claim was valid and infringed.
Interestingly, instead of granting an immediate injunction to prohibit further use of the technology, the Court of Appeal allowed BT a stay of the injunction for two weeks while it changed the way its system operated. ASSAI agreed on the basis of a £250,000 payment per week. This was calculated on a remarkably high royalty rate of 10 per cent of BT’s NGA business, although it will be repayable if the final damages calculation turns out to be lower.
A second recent Court of Appeal case on claim construction was low-tech, by comparison, but provides important guidance on the inclusion of cross-referencing numerals in the claims.
An SEB patent dealing with low-fat frying devices was in dispute. A device was sold by SEB as the Tefal Actifry Fryer. Jarden’s competing Breville Halo Health Fryer had a different design – the heat source was in the lid rather than in the main body of the appliance. A key point in this case was whether numerals from the patent text and diagrams, included in the claims, could influence the meaning of those claims. The judge had cross-referred to numbered points in the patent: "the main body 2", "the lid 2C", "the side skirt 2B", and "the base 2A" to conclude that the phrase "mounted on the main body" included something mounted in the lid.
The Court of Appeal said that this was wrong. The numerals could be considered only as a general guide ("to get the map the right way up") but they should be ignored when working out the meaning of the claim language. Read without the numerals, the term ‘main body’ did not include the lid, so a device with the heater in the lid did not infringe.
For those involved with drafting patent claims it would be worth remembering the advice of the Court of Appeal in an earlier case: "patentees would wisely refrain from inserting numbers in case they were used against them."
A particular problem for the pharmaceuticals industry is the protection of second medical uses. Once the original molecule has already been patented, it is in the public domain. Further research may reveal a second or third application for the medicine but how can that be patented? Methods of medical treatment are expressly excluded from patentability in Europe.
In the past the solution to this problem was to claim the use of the molecule in the preparation of a medicine for the treatment of the second disease – the so-called ‘Swiss-form’ claim. This was superseded in by a new format in 2007 (the molecule for use in the treatment of the second disease) but Swiss-form claims are widespread in older patents protecting products that may now be at the peak of their profitability.
Despite their wide use there have been few decisions across Europe that deal with the interpretation of Swiss-form claims. Now an English judge has concluded that for the claim to be infringed there must be a ‘subjective intent’ on the part of the manufacturer to make the medicine for the patented purpose.
Obvious practical difficulties arise. How would such a subjective intent be proved? And how would a subsequent seller or user of the product know whether the product was infringing? This decision is controversial and seems likely to be appealed. Indeed, in a further decision given two weeks later the judge invited an appeal to the Supreme Court.
Hospira v Genentech
Genentech’s advanced biological breast cancer drug, Herceptin, has been the subject of extensive patent litigation, with Genentech trying to prevent generic launch by Hospira following the expiry of protection for the drug itself.
A decision in November focused on product by process claims. The patents in this case related to formulations of Herceptin. The key claims used the wording "A formulation comprising a lyophilised mixture of…". The judge read this as claiming a product by reference to the process through which it was made – a "product by process" claim.
The European Patent Office has generally treated product by process claims as meaning the product obtained by any method, and has assessed novelty on that basis. The decision in the Genentech case, however, requires the process part of the claim to be taken into account when assessing infringement. So if someone makes the product using a method other than the one claimed, the patentee should not be able to stop him.
Other patents claiming a specified treatment regime for the drug were the subject of a trial last April. The judge felt that the claimed treatment regime would be obvious to pharmacokinetics experts and this has now been confirmed by the Court of Appeal.
Aren’t those claims too broad?
Biopharmaceutical company Idenix has sued competitor Gilead on a patent claiming compounds for the treatment of viral infections such as Hepatitis C. Gilead responded that the patent was invalid claiming much more broadly than was justified by the patent disclosure. The patent was amended to narrow down the hugely broad claims in the application, but claim 1 still covered at least 50 billion compounds.
The decision? The judge felt that the patent did little more than invite the reader to screen a range of compounds for anti-viral activity. The patent did not make a meaningful contribution to the field because it included many compounds that did not solve the technical problem of combatting viral disease - it was invalid for lack of inventive step.
In addition, finding compounds within the claimed field that actually worked would in itself be a research project. The patent did not provide the reader with enough information to produce the claimed materials - it was also invalid for insufficiency.
Threatening letters before the patent is granted
A patent owner becoming aware of infringing activity will generally begin by contacting the alleged infringer. Although it is sensible to begin with an attempt to warn off an infringer, or to negotiate a settlement, this needs to be done with care, because of the danger of being accused of making "unjustified threats". Although currently subject to review following a Law Commission report in April 2014, these UK law provisions often still catch patent holders out.
A recent case looks at how this applies to a patent application, rather than a granted patent.
Civil engineering company Van Den Noort had applied for a patent for a self-closing flood barrier and related flood protection method. Van Den Noort had granted an exclusive licence to Global Flood Defence. Relations cooled and the licence was terminated. Global Flood Defence continued to sell a modified product, arguing that their new version did not come within the patent.
Van Den Noort took action, putting notices on their website and writing to complain of the alleged infringement to a Member of Parliament, who was involved in hosting an event on flood resilience.
When Global Flood Defence sued for threats, Van Den Noort responded that Global Flood Defence’s sales were infringing and so they had a defence. Did it matter that the patent was not yet granted?
The judge decided that the interaction between (a) the rule allowing a patentee to recover damages for activity between the date of publication of a patent application until the date of grant and (b) the rules on threats meant that Van Den Noort’s defence could go forward to trial. The fact that Van Den Noort was making threats about litigation before grant (ie before they were allowed to sue) did not matter. What was important was that Van Den Noort might in future be able to recover for the activity currently taking place.