Antibody Patents in the U.S. & Japan: If You Claim More, You Better Enable More
The Supreme Court of the United States (SCOTUS) recently affirmed lower court decisions invalidating two Amgen patents related to its Repatha products. Around the same time, the Japanese IP High Court ruled a corresponding Japanese patent invalid for failing to meet support requirements in a decision that overruled its own previous decision finding the same patent valid. The take-home message for U.S. and Japanese patent applicants: for functional claims to a broad group of antibodies, be sure to include support for members across the full scope of that group in the application or risk losing your patent rights. As Justice Gorsuch put it in the Amgen SCOTUS decision, “[t]he more one claims, the more one must enable”.
The SCOTUS decision
In the U.S., Amgen obtained two patents which, together, claim “the entire genus” of antibodies that bind to specific amino acids on the naturally occurring protein PCSK9 and which block PCSK9 from binding to and degrading LDL receptors, the body’s blood-stream LDL cholesterol removal mechanism. Amgen sued Sanofi for infringement and, in its defence, Sanofi alleged invalidity of U.S. Patent No. 8,829,165 and U.S. Patent No. 8,859,741 under 35 U. S. C. §112(a) of the Patent Act, which requires that an invention be described in a patent specification “in such full, clear, concise, and exact terms as to enable any person skilled in the art . . . to make and use the [invention]”. Sanofi argued that Amgen’s patents claim millions more antibodies than Amgen is entitled to claim beyond the 26 antibodies it made and disclosed in its patents. Sanofi won before both the district court and Federal Circuit. See our previous article discussing the lower court decisions.
Before the lower courts, Amgen had argued its patents are enabling because they teach scientists two methods to make antibodies according to the claims: the “roadmap” and the “conservative substitution” methods. Sanofi countered that the disclosed methods amount to little more than a guide to a trial-and-error process of discovery and that neither method could enable a skilled person to reliably generate additional antibodies having the same functions as the 26 antibodies described by Amgen.
In the unanimous Amgen SCOTUS decision penned by Justice Gorsuch, the Court reviewed jurisprudence considering the Patent Act enablement requirement, including decisions rendered in patent disputes over Samuel Morse’s telegraphic system, Thomas Edison’s bamboo-based incandescent lamp and the Perkins Glue Company’s starch-based glue. The Court found the jurisprudence “reinforce[s] the simple statutory command. If a patent claims an entire class of processes, machines, manufactures, or compositions of matter, the patent’s specification must enable a person skilled in the art to make and use the entire class. In other words, the specification must enable the full scope of the invention as defined by its claims. The more one claims, the more one must enable”. However, the Court reined this statement in by qualifying that it is not required that a specification “always must describe with particularity how to make and use every single embodiment within a claimed class”, and suggested that a specification that “discloses some general quality running through the class that gives it a peculiar fitness for the particular purpose… may reliably enable a person skilled in the art to make and use all of what is claimed, not merely a subset”. The decision also emphasized that “a reasonable amount of experimentation” required by the skilled person to make and use the patented invention does not necessarily render a disclosure inadequate and that what is reasonable in any case will depend on the nature of the invention and the underlying art.
In applying its legal analysis to the Amgen case, the SCOTUS found that “the claims before us sweep much broader than [the] 26 antibodies” described by Amgen and concluded that, even allowing for a reasonable degree of experimentation, the full scope of the asserted claims lacked enablement and are therefore invalid. The Court rejected Amgen’s argument that its “roadmap” and “conservative substitution” methods are enabling, finding them to amount to “little more than two research assignments” that would force a scientist to engage in “painstaking experimentation” to find out what works.
The Japanese IP High Court decision
A few months before the Amgen SCOTUS decision was rendered, the Japanese IP High Court handed down a similar decision in a closely related matter. At issue were the claims of Amgen’s Japanese Patent No. 5,705,288, which encompass any monoclonal antibody capable of neutralizing the binding of PCSK9 to LDL receptors and which competes with the 21B12 antibody as defined in the specification for binding to PCSK9. Similarly to the U.S. Amgen patents described above, the ‘288 patent discloses screening methods to obtain antibodies that meet the claimed requirements and discloses some examples of antibodies exhibiting the claimed neutralizing function. Sanofi had previously lost a validity attack on the ‘288 patent before the IP High Court and was ordered to withdraw its Praluent products from the market in Japan. Regeneron, a co-developer of Praluent, then launched its own validity attack against the same patent.
In the Regeneron case, in considering whether the support requirement under the Patent Act is met for the ‘288 patent, the IP High Court found the specification disclosed no explanation of how an antibody could neutralize binding of PCSK9 to LDL receptors and therefore concluded that the specification does not support a claim to “any” antibody which both competes with the 21B12 antibody and neutralizes PCSK9 to LDL receptor binding. The ‘288 patent was invalidated. The Court’s comments suggest that inclusion in the specification of a mechanism explaining how the claimed antibodies carry out the neutralizing function may have been enough to save the ‘288 patent. The Court also remarked that it came to a different conclusion from its previous decision regarding validity of the same patent on the basis of new arguments and evidence that were presented.
The U.S. and Japan Amgen decisions may be indicative of a narrowing global trend set to work against life science patentee interests. In both cases, the claimed antibodies were defined based on ability to bind defined targets and functional outcomes from such binding. Both courts placed more emphasis on the scope of the claims rather than what the skilled person could determine from the specification. As forecasted in our previous article, going forward, functional claims in U.S. life sciences patents may become held to a more rigorous enablement standard compared to functional claims in, for example, patents over mechanical, IT or computer-implemented technologies. The same may also be said for Japanese life sciences patents.
Functional claims to antibodies are allowable in other major jurisdictions such as Europe, South Korea and Australia, which is permissive for strong patent rights as they are more difficult to design around compared to structurally defined claims. For example, in Canada, the enablement requirement is fulfilled for antibody claims if the skilled person would be enabled to produce an antibody according to the claims without displaying inventive ingenuity or undertaking undue experimentation. It remains to be seen whether the U.S. and Japanese Amgen decisions will influence how life science patent enablement is assessed before Canadian courts.
At least for U.S. and Japanese life science patents, applicants are strongly advised to include support for both broad and narrow embodiments in their applications to provide fallback positions during prosecution and protect against post-grant validity attacks. For antibody patents, this could include description of the invention from both a functional perspective (e.g., describing binding characteristics or effects on downstream activity markers) and a structural perspective (e.g., describing nucleic or amino acid sequences of the antibody complementarity-determining regions or CDRs), accompanied by as many actual or prophetic examples as available that could be relied upon to demonstrate enablement across the full scope of broad claims. When pursuing functional claims, it would also seem advisable to include mechanistic information corresponding to functional claim limitations in the specification. At least for Japan, including explanation of the mechanism for how the claimed antibodies carried out their neutralizing function may have saved the patent. It is interesting to consider whether inclusion of such information would have impacted the outcome in Amgen.