The Federal Court has dismissed an appeal by Sanofi against a decision of the Australian Patent Office (APO) to allow five Amgen patent applications containing functional antibody claims directed to Amgen’s cholesterol-lowering drug Repatha® (evoloumab) to proceed to grant.
The favourable decision for Amgen in Sanofi v Amgen Inc. (No 3) [2025] FCA 387 (Sanofi) is the latest instalment in a long-running worldwide dispute over the Repatha® patents – including in the US and Europe, where corresponding patents were found to be invalid for lack of enablement and inventive step, respectively.
Relevantly, Sanofi was decided under Australia’s pre-Raising the Bar (pre-RTB) law, which sets a much lower threshold for sufficiency and support than the current European-type standard. Therefore, the Federal Court’s findings have limited applicability to applications and patents subject to the RTB sufficiency and support requirements.
Background
The five patent applications at issue (Applications),[i] each being divisionals of the same parent application and having the same specification, are titled “Antigen Binding Proteins to Proprotein Convertase Subtilisin Kexin Type 9 (PCSK9)”. As the title suggests, the Applications relate to antibodies that inhibit PCSK9, which is involved in degradation of low density lipoprotein (LDL) receptors on the surface of the liver. Degradation of LDLR receptors (LDLR) reduces the body’s ability to remove circulating LDL-C (“bad cholesterol”) from the blood, thereby increasing the risk of cardiovascular disease.
The specification exemplifies several antibodies that bind to PCSK9 and block its binding to LDLR, including 31H4 and 21B12 (the Reference Antibodies), and antibodies that compete for binding with the Reference Antibodies. The specification also includes a crystal structure of PCSK9 ProCat domain bound to fragments of the Reference Antibodies, suitable methods for generating and testing for antibodies having the claimed functions, and suitable competition assays.
The Applications include the following functional claim types:
- Epitope Claims, which define the antibody by its ability to bind an epitope on PCSK9 and its ability to block binding of PCSK9 to LDLR;
- Residue Claims, which define the antibody by its ability to bind to certain amino acid residues and its ability to block binding of PCSK9 to LDLR; and
- Competition Claims, which define the antibody by its ability to compete for binding to PCSK9 with the Reference Antibodies (defined by their heavy and light chain sequences).
Sanofi opposed the grant of patents on the Applications before the APO on the grounds, inter alia, that the claims fail to define the invention, lack fair basis, are not fully described, and lack inventive step. Sanofi’s arguments centred around the failure of the claims to include any structural features to define the antibodies of the invention.
The APO dismissed the opposition in its entirety in Sanofi v Amgen Inc. [2022] APO 67, and Sanofi appealed the APO’s decision to the Federal Court.
Appeal decision
On appeal, Sanofi argued that the claims did not properly define the invention because none of the Epitope, Residue and Competition Claims sufficiently defines the structure of the antibody (e.g., by reference to the amino acid sequence of its variable domain).[ii] However, Nicholas J was not persuaded that it was necessary to define the antibody by its amino acid sequence to define the invention.[iii] In that regard, his Honour agreed with Amgen that it was not necessary to know in advance whether a product will fall within the scope of the claims (only that it can be ascertained after the product has been made).[iv]
In relation to full disclosure, which merely requires that the skilled person be able to “produce something within each claim without new inventions or additions or prolonged study of matters presenting initial difficulty”,[v] Sanofi argued that each claim related to more than one invention (being a large number of individual antibodies with its own epitope on PCSK9) and, as such, it was necessary to describe how to produce something pertaining to each invention.[vi] However, because the specification enabled the skilled person to generate at least one antibody within each of the claims (e.g., the Reference Antibodies), Nicholas J was satisfied the requirement for full disclosure was met.
In relation to fair basis, which requires the specification to contain “a real and reasonably clear disclosure” of the invention,[vii] Nicholas J was essentially satisfied this requirement was met by the inclusion of consistory statements in the description corresponding to the language of the claims together with the exemplification of antibodies that interact with and neutralise PCSK9, and that compete for binding with the Reference Antibodies.
As to inventive step, the relevant question was whether a skilled person “would have been directly led to try to generate such antibodies as a matter of course in the expectation that they may well block or inhibit binding between PCSK9 and LDLR”.[viii] Based on the evidence, Nicholas J considered “the pathway to the generation of anti-PCSK9 antibodies was not clear or straightforward and was the subject of considerable uncertainty relating to PCSK9’s mechanism of action … and whether it would even be possible to generate an antibody to PCSK9”.[ix] Accordingly, his Honour was not satisfied that the skilled person would have directly been led as a matter of course to try an anti-PCSK9 antibody with the requisite expectation of success.
Comparison to US and European proceedings
In 2023, the US Supreme Court unanimously upheld the lower court’s decision to invalidate claims corresponding to the Epitope and Residue Claims for lack of enablement.[x] In the US, enablement requires the specification to include “a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art … to make and use the same …”.[xi] In particular, the US Supreme Court considered that a skilled person would be required to undertake “painstaking experimentation” to make the “entire universe of antibodies” within the scope of the claims.
In Europe, the European Patent Office (EPO) and the Unified Patent Court (UPC) have reached opposing decisions in relation to European Patent No. 3666797 (EP ‘797), which claims antibodies that “bind to the catalytic domain of a PCSK9 protein of the amino sequence of SEQ ID NO: 1, and prevents or reduces the binding of PCSK9 to LRLR” for use in treating or preventing hypercholesterolemia or an atherosclerotic disease, or reducing the risk of a recurrent cardiovascular event, related to elevated serum cholesterol levels.[xii]
Earlier this month, the EPO rejected in its entirety an opposition by Sanofi to EP ‘797, including on the grounds of sufficiency and inventive step. In contrast, in its first ever revocation decision in 2024, the UPC found that the claims of EP ‘797 lacked an inventive step.[xiii] The UPC did not address any other grounds of invalidity, including sufficiency. Both decisions have been appealed, with the UPC Court of Appeals’ judgement expected to be handed down in early May 2025.
Would the Raising the Bar law yield a different decision?
In Australia, the decision – at least insofar as it relates to internal validity (i.e., full disclosure and fair basis) – would likely have been very different if the decision had been made under the RTB law, which applies to all applications and patents for which examination was requested on or after 15 April 2013. Under the RTB law, the internal validity requirements include that:
- the specification provides a clear enough and complete enough disclosure to enable a skilled person to perform the claimed invention without undue burden or further invention (Sufficiency Requirement);[xiv] and
- the claims are supported by matter disclosed in the specification, such that there must be a basis in the description for each claim and the scope of the claims must not be broader than is justified by the applicant’s contribution to the art (Support Requirement).[xv]
While there have not yet been any post-RTB judicial decisions relating to functional antibody claims, the APO provides some guidance in relation to the allowability of certain antibody claims post-RTB, including:[xvi]
[A claim] to an antibody that specifically binds epitope A would be fully enabled where the applicant had disclosed the epitope and shown that antibodies can be raised against it. This is because raising antibodies to that epitope is a principle that can be generally applied to produce antibodies over the whole scope of the claim. The specification need only disclose one such antibody.
However, recent case law suggests that relying on a principle of general application can be dangerous (see, e.g., our previous article), particularly in the event of certain later selection inventions within the scope of the claims.
In the present context, reliance on a principle of general application may be unsafe where antibodies falling within the scope of the claims are later found to be unable to satisfy a particular functional requirement. For example, post-RTB, the Competition Claims may be at risk of (retroactively) failing to satisfy the Sufficiency and Support Requirements if competing antibodies were later found that are unable to block binding of PCSK9 to LDLR.
Notably, the APO often objects to competition claims on the basis that a competing antibody is not guaranteed to bind the same, defined epitope to which an exemplified antibody binds. The APO takes the position that where a competing antibody binds to an “overlapping” epitope that does not share the same functional properties as a defined epitope, an undue burden of experimentation would be required to identify competing antibodies with the same functional requirements as an exemplified antibody. It remains to be seen whether the Federal Court will adopt a similar approach to the APO.
While the test for inventive step has remained largely the same post-RTB,[xvii] in view of the growing propensity for the Australian courts to follow European law, it will be interesting to see how the EPO Board of Appeal and the UPC Court of Appeal tackle the issue of inventive step.
Concluding remarks
The Sanofi decision is good news for functional antibody claims in Australia. However, the higher threshold for sufficiency and support means that such claims may pose a greater risk post-RTB. In view of the potential drawbacks of relying on genus claims that define antibodies by functional features, it is generally recommended to include claims that structurally define the exemplified antibodies in Australia.
In view of the ongoing nature of this dispute globally, it would not be surprising if Sanofi appeals the Sanofi decision to the Federal Court, e.g., on the basis of inventive step. We will keep you informed of further developments in relation to this matter.
[i] Australian Patent Application Nos 2013203677, 2013203748, 2013203685, 2013203689 and 2013203751
[ii] Sanofi at [193]
[iii] Sanofi at [218]
[iv] Sanofi at [196]
[v] Kimberly-Clark Australia Pty Ltd v Arico Trading International Pty Ltd (2001) 207 CLR 1 at [25]
[vi] Sanofi at [456]-[457]
[vii] Lockwood Security Products Pty Ltd v Doric Products Pty Ltd (2004) 217 CLR 274 at [69]
[viii] Sanofi at [426]
[ix] Sanofi at [426]
[x] Amgen v. Sanofi, No. 21-757, concerning US Patent Nos. 8,829,165 and 8,859,741
[xi] 35 U.S.C. §112(a)
[xii] Parent European Patent No. 2215124 was upheld by the EPO Board of Appeals in 2020, but limited to certain sequences, which allowed Sanofi to market its Praluent® product
[xiii] Sanofi v Amgen, UPC_CFI_1/2023
[xiv] Patents Act 1990 (Cth), s 30(2)(a), taken together with the Explanatory Memorandum accompanying the Intellectual Property Laws Amendment (Raising the Bar) Bill 2011 (EM), Item 8
[xv] Patents Act 1990 (Cth), s 30(3), taken together with the EM, Item 9
[xvi] IP Australia Patent Manual of Practice and Procedure, Section 5.6.7.8 Annex A
[xvii] Noting, however, that the pre-RTB requirement that a prior art document be “ascertained, understood and regarded as relevant” for the purpose of assessing inventive step has been removed