In brief

The US Supreme Court has weighed in on the long-running litigation between Amgen and Sanofi regarding Amgen’s patents for antibodies engineered to reduce levels of low-density lipoprotein (LDL) cholesterol. In finding the patents invalid for lack of enablement, the Court provided guidance for when patentees may be able to stake a claim to a broad genus or class, and where too much work is left to a person skilled in the art (PSA) for the patent bargain to be met.

This decision is unlikely to impact the related proceedings currently before the Federal Court of Australia, which were brought under Australia’s “old” Patents Act. It does, however, provide insight into how broad functional antibody claims may be treated under the “new” Raising the Bar regime which introduced a similar “enablement” requirement.

Background

High levels of LDL cholesterol are linked to cardiovascular disease, heart attacks and strokes. Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a naturally occurring protein that binds to and facilitates the degradation of LDL receptors (LDLRs) responsible for extracting LDL cholesterol from the bloodstream. Elevated levels of PCSK9 leads to increased degradation of LDLRs, resulting in less LDL cholesterol being cleared from the circulation (and, hence, higher levels of circulating LDL cholesterol).

Amgen and Sanofi each developed its own anti-PCSK9 antibody (marketed as Repatha and Praluent, respectively), which are able to bind to PCSK9 and thereby block PCSK9 from binding to LDLR. In 2011, Amgen and Sanofi each filed patents based on the unique amino acid sequences specific to their respective antibodies.

In 2014, Amgen filed two additional related patents, US patents number 8,829,165 (US 165) and 8,859,741 (US 741) (together, the Amgen Patents), each of which claims a class (or genus) of antibodies. US 165 claims antibodies which bind to at least one of 15 specific amino acid residues of PCSK9 and block PCSK9 from binding to LDLR. US 741 claims antibodies which bind to an epitope on PCSK9 comprising at least one of two specific amino acid residues of PCSK9 and block PCSK9 from binding to LDLR. The Amgen Patents also describe two methods that a PSA could use to make other antibodies that fall within the scope of these claims – the roadmap method and the conservative substitution method.

District Court and Federal Circuit Court decisions

These proceedings have a long history, beginning with infringement proceedings commenced by Amgen against Sanofi in 2014, soon after the Amgen Patents were granted. Sanofi admitted infringement but asserted that the claims of the Amgen Patents were invalid, including for lack of enablement.

Amgen was initially successful before the District Court until the United States Court of Appeals for the Federal Circuit remanded for a new trial, including in relation to the District Court’s evidentiary rulings and jury instructions regarding Sanofi’s invalidity arguments. On the second occasion before the District Court, Sanofi obtained judgment as a matter of law for lack of enablement. Amgen appealed to the Federal Circuit, which upheld the ruling.

In November 2021, Amgen petitioned the United States Supreme Court which was granted in November 2022.

Supreme Court decision

On 18 May 2023, in a joint opinion delivered by Justice Neil Gorsuch, the Supreme Court was unanimous in concluding that the Federal Circuit was correct in finding that the Amgen Patents were invalid for lack of enablement.

Central to the debate was that the Amgen Patents disclosed the amino acid sequences of 26 antibodies which bind to PCSK9 but the claims covered a “vast” class of antibodies defined by their function (i.e. binding to particular areas of PCSK9 and blocking PCSK9 from binding to LDLR).

Amgen argued that the claims were enabled because the PSA could make and use every functional antibody if they followed either the roadmap method or the conservative substitution method. The Court however was not convinced, finding the roadmap method was merely a step-by-step description of the trial-and-error Amgen had taken in identifying functional antibodies. As to the conservative substitution method, the Court found that this was an uncertain prospect given the state of the art was that changing even one amino acid in the sequence can alter the antibody’s structure and function, and the effect of substitutions could not always be accurately predicted.

Amgen also made three alternative arguments in support of its case. First, Amgen argued that the Federal Circuit had erred by conflating the question of whether an invention is enabled with the question of how long it may take a PSA to make every embodiment within a broad claim. This argument was rejected.

Secondly, Amgen argued that there is only one universal enablement standard whereas the Federal Circuit had applied a higher standard to the present style of claims which encompass an entire class of embodiments defined by their function.

The Supreme Court agreed in principle with Amgen regarding the one statutory enablement standard, but considered the Federal Circuit’s application of the standard was entirely consistent with Congress’s directive and Supreme Court precedents.

Thirdly, Amgen relied on a public interest ground, asserting that a ruling against it risked destroying the incentives to innovators that lead to breakthrough inventions. The Supreme Court was not convinced, stating this is a matter for Congress and its role was to apply Congress’ requirement for enablement.

In reaching these conclusions, the Supreme Court made clear that a patent specification need not describe with particularity how to make and use every single embodiment within a claimed class. For instance, it may suffice to give one or more examples if the specification also discloses “‘some general quality … running through’ the class that gives it ‘a peculiar fitness for the particular purpose’”. Nor is a specification necessarily inadequate just because it leaves the PSA to engage in some measure of adaptation or testing, although what is a reasonable amount of experimentation will depend on the nature of the invention and the underlying art. A specification must enable the full scope of the invention as defined by the claims, so that the public has the full benefit of the invention after expiry of the patent term.

Related Australian proceedings

As we have reported previously, Amgen and Sanofi have been engaged in a dispute regarding related Australian patent applications for a number of years. Five of Amgen’s patent applications were opposed by Sanofi, with the Australian Patent Office finding all claims to be valid. The dispute is now on appeal before the Federal Court, with closing submissions heard in December this year.

As discussed in our previous article, the issues raised before the US Supreme Court will not arise in Australia, as the opposition is under the Patents Act as it stood prior to amendments made by the ‘Raising the Bar’ regime, which introduced the new requirements of support (replacing fair basis) and clear and complete disclosure (replacing sufficiency). The Raising the Bar regime set a higher standard for enablement and support and aligned Australian patent law closer to the corresponding requirements in the UK, Europe and to a lesser extent the US. Having regard to the Full Federal Court’s recent reliance on UK authorities in relation to clear and complete disclosure,^[1] so too may the Federal Court have regard to the US Supreme Court’s guidance on antibody claims.

Considerations for the future

Over the past decade, biologic medicines, including antibodies, have taken over the top rungs of the highest selling medicines by value across the globe. While patent disputes regarding small molecules typically involved litigation between “innovator” and “generic” pharmaceutical companies, for biologic medicines there are often multiple innovators each with their own (different) antibody to the antigen of interest (e.g. antibodies which target different regions or amino acids of that protein). This change in landscape, as a practical matter, encourages a measure of discipline to the breadth of claims to biologic medicines.

Pharmaceutical patents have traditionally claimed a large genus of compounds, at least in part due to the more patentee-friendly pre-Raising the Bar regime. Patentees seeking to stake such broad claims in relation to biologic medicines, in particular by way of claims by function rather than structure, will however face challenges under the Raising the Bar regime. As the US Supreme Court’s decision makes clear, patentees will need to take care that they do not cast their net too widely for fear of losing the entire catch.