The US Supreme Court’s refusal to examine the written description requirement set out by the Federal Circuit has renewed doubts over the value of patents in the increasingly lucrative antibody space, and prompted drug companies to find alternative protection avenues.
Senior sources at Novartis, Bristol-Myers Squibb (BMS) and other drug firms say there has been considerable uncertainty over whether patents could offer sufficient protection for antibodies since the 2017 Amgen v Sanofi ruling and the subsequent application of Section 112 under Title 35 of the US Code.
The judgment reaffirmed that Section 112, which sets out patent specification law, has enablement and written description requirements. The latter requirement, known as the possession test, makes it difficult to write patent specifications for antibodies that sufficiently demonstrate possession while ensuring that the invention is not easy to design around.
Some drug companies had hoped that a higher court ruling might uphold a special possession test for antibodies, create a new one or reject the posession requirement altogether in favour of a Section 112 standard that only required enablement or at least create more certainty in the industry. But those were dashed at the beginning of this year when SCOTUS declined to hear an Amgen v Sanofi appeal.
“Anyone who has relied on the old newly-characterised antigen test in drafting their patent over the past 15 years is now likely to have their patent invalidated because the Federal Circuit will have the last say,” says Corey Salsberg, global head of IP affairs for Novartis in Washington DC.
He explains that while me might see some drug firms that claimed numerous antibodies under the previous Section 112 characterisation test trying to bring said test back by asking SCOTUS to take their case, any case similar to Amgen's is unlikely to be heard by the court and will ultimately be judged in a similar manner by the Federal Circuit.
Salsberg says that while not all companies agreed with Amgen’s petition to SCOTUS to get rid of the written description requirement altogether, many wanted to see a revision of the case law, or at least get clarity on what is required for antibody claims.
“Most acknowledge that there are two sides to the matter; the new test is creating uncertainty because it may make numerous antibody patents invalid and is making it difficult for companies to claim what they invented with sufficient scope so that others cannot come in and take the fruits of their labour.
“But it is also a fair point that the old test enabled inventors to cover every conceivable antibody based on the characterisation of an antigen, which may be too broad, particularly where the competitive product is a different medicine. In other words, the Amgen v Sanofi case was about how much breadth should be given to to these types of antibody claims, and whether the fruits of separate innovative efforts should really be precluded by a patent that, while broad, covers a different medicine.”
The uncertainty surrounding Section 112 is particularly damaging to the antibody innovator industry and the biopharmaceutical sector broadly because of the medicinal and financial opportunities offered by the technology.
“We have created a system that does not reward pioneers for their antibody inventions,” says Henry Hadad, deputy general counsel at BMS in New York. "Particularly against other branded companies that are able to generate antibodies to the same target but avoid patent infringement due to the current application of Section 112.” He adds that the discovery of broad therapeutic applications of antibodies against a particular target are often the work of academic institutions and small companies, which have been instrumental in the fight against cancer and other serious diseases.
"I would say that the SCOTUS decision to not hear Amgen v Sanofi means its is unlikely that the courts will reverse the current trends for antibody patents."
These innovations require huge investments with uncertain outcomes that should be encouraged through more robust patent protection.”
One industry source adds that his company enforced its antibody patents against another large drug company in Europe with great success, but did not try the matter in the US because of its case law.
The assistant general counsel at a pharmaceutical company adds: “Right now all we have is negative – the courts are saying that this is the test and you can’t claim all antibodies by characterising one antigen.
“But I’m not sure there is any company saying what the test should be – it is practically difficult because there are so many antibodies that combine with the same antigen.” The antibody production sector is expected to be worth $26.16 billion by 2016, according to research by market intelligence firm Reports and Data published in May 2019. The report suggests that this growth can be attributed to the increasing prevalence of contagious diseases and demand for protein therapeutics.
Alternative protection
With the absence of certainty when it comes to the new possession test in the US, sources say epitope mapping and creation and screening of antibodies that bind to a target antigen have become important ways of getting sufficiently broad patent protection.
Before the Amgen ruling, the Federal Circuit allowed Section 112 to be satisfied by the antigen characterisation test, enabling patent attorneys to describe the structure of an antigen in a claim and get all corresponding antibodies that combine to it.
The new test requires attorneys to demonstrate through the specification that the business has invented what it says it has invented, which means attorneys must describe the representative number of species in a genus.
The first problem with that test for antibodies is that there may be hundreds of antibodies, and while it is not practical to list them all, it is uncertain how many would be considered representative.
The second is that it is easy to make a small structural change to an antibody to design around a claim; which means businesses risk losing out on a valuable antibody because they failed to list it in the specification among all the others.
“Making monoclonal antibodies has becomes almost a commodity now that biosimilars have come on the market,” explains the pharmaceutical assistant general counsel. “It used to be that the value was in the process – because you could not come up with the same product if you could not do the same process. But now the product itself it the valuable output.
“Technology has now matured to the point that any company that wants to make antibodies can do it – even with a different and perhaps inferior process,” she adds.
Salsberg at Novartis says that companies are now looking at other ways of satisfying the Section 112 requirement for antibodies, such as focusing on complementarity determining regions (CDRs). Defining and claiming functional or structural similarities or specific sequences within these CDRs is one approach, he says, because there may be protein sequences that are common that a company can either claim directly or disclose enough to support a broader claim.
He warns, however, that claiming directly for such sequences only works when a business is able to characterise a number of antibodies that reflect those sequences.
“There are lots of different approaches but these are some of the ideas being floated around. In some cases it might be possible to characterise structures of an antibody itself.
“The assumption is we are in this world of showing through Section 112 the representative number of species or common structural features. The challenge becomes working out if there is a way to identify those features or disclose the representative number of species,” Salsberg concludes.
Sources say that another option available to drug companies is to claim the binding of the antigen by epitope and the function. The problem with this approach is that it would require a representative number of examples of the genus; a figure which remains unknown.
"These points raised are a fair discussion on the considerations around antibody patenting," says Hadad at BMS.
But he says that the possession test is a court doctrine that can ultimately be remedied by the courts. but adds that it is unlikely that the Federal Circuit or SCOTUS will positively address this issue at this point.
“The current case law can cut both ways for biopharmaceutical companies– but the big picture is that we want to encourage pioneers and help patients by discovering new therapies. If companies want to copy antibody innovations, they should pay their fair share.”
Antibodies are immensely important in the biopharmaceutical space, and yet companies are having a hard time protecting them in the US. No one is sure what a better test that sits somewhere between possession and characterisation might be.
Certainly, the US courts won’t be looking at the matter any time soon, and the onus sits with patent attorneys to find new and clever ways to protect these lucrative inventions.
