Best Practices for Conquering the Enablement Requirement After Amgen

The Supreme Court’s decision in Amgen Inc. v. Sanofi, 143 S. Ct. 1243, 1248 (2023), found that antibody claims defined by their binding and blocking function lacked enablement. As noted in the decision, the specification identified 26 antibodies by their structure. It also provided a method of screening antibodies by generating candidate antibodies and screening for their binding properties, and a method of conservative substitution, which called for modifying an antibody with a known function by replacing select amino acids in the sequence and then testing the result for its function. Despite this extensive disclosure, Amgen “failed to enable all that it has claimed, even allowing for a reasonable degree of experimentation.” Id. at 1246.

The Supreme Court’s decision relied highly on the unpredictability of the art—scientists cannot always predict how substitutions of even single amino acids will affect the binding and/or blocking function of an antibody. The disclosed methods of identifying species within the scope of the claim were little more than “trial and error.” Id. at 1257. More disclosure was required due to the breadth of the claim: “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.” Id. at 1254.

The consensus seems to be that Amgen generally forecloses broad genus claims in the biotechnology field. The remaining questions are:

• Are patent applicants limited to claiming species (e., sequence listings)?
• What steps may a patent applicant take to obtain broader coverage?

Functional Claims are (Mostly) Out

The Supreme Court clearly disapproved of functional genus claims, particularly in the antibody context, but did not foreclose all genus claims, noting that a few examples might support a genus claim if a “general quality” is disclosed that “may reliably enable a person skilled in the art to make and use all of what is claimed, not merely a subset.” Amgen, 143 S.Ct. at  1255. Some experimentation is permissible if the amount of experimentation is reasonable. Amgen, however, does not provide a lot of clarity on what types of claims would be acceptable, or what type of “general quality” would be sufficient to satisfy the enablement requirement.

How Would You Support a Valid Functional Claim or a Broad Genus?

In the antibody context, the Supreme Court has stated “scientists understand that changing even one amino acid in the sequence can alter an antibody’s structure and function” id. at 1249, and, in his amicus brief, Sir Gregory Paul Winter, Nobel Prize-winning molecular biologist, stated “an antibody scientist is unable to predict the function of an antibody from its sequence.” It is therefore difficult to contemplate a “general quality” common to antibody amino acid sequences. There are deep learning methods being developed to predict antibody interactions based on sequence, but these methods are not yet at the stage to assist in patent procurement.

What to Claim Now?

A patent application should include lots of species specified by their amino acid sequence or structure. These claims can include, with respect to antibodies, sequences for the heavy chain and light chain CDR (complementarity-determining regions), and heavy and light chain variable regions, with (limited) substitutions. At this point, it is difficult to say whether the use of claims with a percentage of sequence identity (90% identity with SEQ ID NO: 1) or conservative substitutions will satisfy the enablement requirement, particularly if the claim is challenged in litigation. As noted in The Antibody Patent Paradox, such claims will not cover species with similar binding if they have different structures, and may be invalid if many of the species are inoperative or require further screening. 132 Yale L.J. 994, 1054 (Feb. 2023). Yet, if the number of substitutions is limited, the examiner will likely allow the claim.

Patent owners should consider reissue if they don’t have a current patent or pending application with sequence claims.

For small organic molecule claims, the core structure will usually have variable substituents (usually represented by an R). The potential variations at each location should be limited to those with common properties. As a back-up, have different groups of substitutions (one small group, one larger group) possible at each location. Provide working (or at least predicted) examples of substitutions at each location.

Some attorneys suggest, in claiming antibodies, to not describe the targeted epitopes because their disclosure likely doesn’t give enough structure to the claims, and would be better maintained as a trade secret. However, patents have been issued recently where the epitope has been specified by sequence listing and the epitope has been mapped to an antibody.

Another suggestion has been to use means plus function claims, which has been successful in certain (non-antibody) applications. Under current case law, an element in a claim may be expressed as a means for performing a function, but is limited to the means described in the specification, and their equivalents. Notably, later-developed equivalents are generally covered. While these types of claims have not been used in the antibody context, they are perhaps worth exploring, as covered in The Antibody Patent Paradox, 132 Yale L.J. 994, 1055-1060 (Feb. 2023).

Enforcement

If you have patented only species claims, are you able to enforce the claims against competitors who have attempted to “design around” your sequence? The answer is unclear. In its brief as amicus curiae, the Solicitor General suggested the “doctrine of equivalents” may apply in enforcing sequence claims, including against a “competitor who has made ‘unimportant and insubstantial changes and substitutions’” to attempt to avoid the patent. No. 21-757 at 32 (Feb. 10, 2023) (citing Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 607 (1950). If the patented and allegedly infringing antibodies bind to the target epitope at different residues and have a different sequence, do they achieve the same result “in the same way”? The use of the doctrine of equivalents relating to inventions disclosed by a sequence listing is an area that should be explored.

Final Note

Amgen addressed the enablement requirement but don’t forget written description. For example, in the recent case of In re Entresto Sacubitril/Valsartan Patent Litig., MDL No. 20-2930-RGA (D. Del. July 27, 2023), the district court confirmed that later-existing technology need not be enabled. As construed, a genus claim encompassed complexes of valsartan and sacubitril. Because complexes were not known in the art as of the priority date, they were “later existing technology that need not be enabled.”  Id. at *62. But, because complexes of valsartan and sacubitril were not described in the specification, the patent was invalid for lack of written description.

Get more tips from the author and other panelists at IPWatchdog LIVE on September 19.

Image Source: Deposit Photos
Author: buchachon_photo
Image ID: 13043324