Case Study: Recently Granted Epitope-Based Antibody Patents in the United States, Europe and Japan
June 13, 2020
“Patent strategies for antibody medicines should include obtaining an antibody sequence-based patent covering the developed drug, in addition to an epitope-based patent.”
Patents involving antibody medicines (antibody patents) are largely grouped into patents specified by antibody amino acid sequences (antibody sequence-based patents) and those not (non-sequence-based patents). Table 1 shows the features of each group. The titled “epitope-based antibody patents” are included in the “non-sequence-based patents”. The epitope is a specific structural unit that resides in an antigen and is recognized by an antibody.
Non-sequence-based patents have a broad scope and are thus very useful for protecting antibody medicines. For instance, in 2014, Bristol-Myers Squibb and Ono Pharmaceutical Company instituted litigation against Merck in countries including the United States, Europe and Japan for marketing Keytruda (an anti-PD-1 antibody), claiming infringement of their non-sequence-based patent on Opdivo (an anti-PD-1 antibody, U.S. Patent No. 8,728,474 for a method of treating cancer). Opdivo and Keytruda have distinct antibody amino acid sequences, but share an antigen and an indication, which are covered by the patent. This litigation ended in a settlement, in which Merck agreed to a payout and royalties.
Meanwhile, in 2014, Amgen instituted litigation against Sanofi in countries including the United States, Europe, and Japan for Sanofi’s marketing of Praluent (an anti-PCSK9 antibody), for the infringement of non-sequence-based patents on Repatha (an anti-PCSK9 antibody) (e.g., U.S. 8,829,165: a patent involving a binding site; U.S. 8,859,741: a patent involving an epitope; JP 5,705,288: a patent involving competition). As a result, Sanofi’s Praluent was discontinued in May 2020, in Japan. The U.S. suit is still in progress. The critical issue in Amgen v. Sanofi (Fed. Cir. 2017) in the United States involved the claims of the Repatha patents (U.S. 8,829,165 and U.S. 8,859,741) characterized by the antibody-binding site (or epitope) in an antigen (the claims are what is called functional claims). Upon Amgen v. Sanofi, on February 22, 2018, the USPTO issued a memorandum to patent examiners clarifying the written description requirement of 35 USC § 112(a) for claims drawn to antibodies. This memorandum stated, as a guideline, that “In view of the Amgen decision, adequate written description of a newly characterized antigen alone should not be considered adequate written description of a claimed antibody to that newly characterized antigen, even when preparation of such an antibody is routine and conventional”.
After this decision and the memorandum were issued, the USPTO standard of written description requirement for functional claims to cover antibody inventions became stricter.
Search for Epitope-Based Antibody Patents
Here, I investigated a recent trend in antibody patents characterized by an antibody-binding site in an antigen. In this article, patents characterized by an antibody-binding site are referred to as “epitope-based patents” for convenience. Not only substance claims (“An antibody….”), but also treatment method claims (“A method for…”) were included in search subjects as long as an antibody binding site was set forth.
Table 2 shows the number of epitope-based patents hit by patent screening under retrieval condition 1. For comparison, the number of antibody-related patents hit under retrieval condition 2 is also shown.
Although the number of antibody patents in the United States was about twice larger than those in the other countries, the number of epitope-based patents in the United States was smallest. This indicates that it is much harder to obtain an epitope-based patent in the United States than in the other countries.
Tables 3 to 5 list epitope-based patents registered from January 2019 to April 2020 in the United States, Europe, and Japan. For comparison, the type of patent claims of the corresponding family patents in the other countries is also provided. Regarding the U.S. patents, the date of patent registration and the category of claim are designated in parentheses.
The number of epitope-based patents allowed in all of the United States, Europe, and Japan is two (EP 2918678, JP 6564408), which is very low. This indicates that different countries have different examination standards and qualities.
When I looked into file wrappers of the seven U.S. patents in Table 3, the examiner pointed out in the office action that U.S. 10,316,105, U.S. 10,221,239, and U.S. 10,196,628 did not comply with written description requirements for functional claims (epitope-based claims), but not issued in U.S. 10,596,245, U.S. 10,442,859, and U.S. 10,259,868, U.S. 10,189,900.
In U.S. 10,221,239, the examiner stated, in Non-Final Rejection (2017-10-12) and Final Rejection (2018-06-14), the claims were rejected because of a lack of written description (In the Final rejection, the examiner cited Amgen v. Sanofi (Fed. Cir. 2017)). As a counterargument, the patentee amended the claims (the range of the antibody-binding site is made narrower) and filed the Remarks (“Applicant therefore notes that the specification provides an antibody within the scope of the claimed limited genus of antibodies. Further, the epitopes recited in claim 1 (SEQ ID NOs: 11-12) are characterized by amino acid sequence and tertiary structure (see Figure 4).”). This amendment and Remarks resolved the claim rejection.
When I looked into file wrappers of the European and Japanese patents (Tables 4 and 5), such a strict standard as in the United States was not found. In Japan, however, the office action was issued because the range of antibody-binding site in the antigen was too wide or inappropriately defined (i.e., support/enablement requirements are not met) in many cases (e.g., JP 6,663,747, JP 6,564,408, JP 6,548,696). The patentees argued by providing the Remarks (e.g., the antibody set forth in the claims was able to be obtained without undue experimentation) and the amendment of the claims (e.g., the range of the antibody-binding site was made narrower). While the number of epitope-based patents allowed was larger in Japan than the United States and Europe, the Amendment and Remarks were often critical.
Hard to Predict
It is not easy to obtain an epitope-based patent, but it is significant to file an epitope-based patent application because competing products can be within the scope of the patent right. Regarding the representative species test and the structural features test used to judge the written description requirement in the United States, the USPTO and CAFC have not indicated to what extent examples should be disclosed specifically. Thus, the situation is that patentability of epitope-based patent applications and patent stability are hard to predict. Meanwhile, it seems to be important to provide, via examples in the specification, as many functional antibodies (species) as possible and to set an appropriate range of antibody-binding site (a wider range of antibody-binding site is likely to result in a rejection).
In addition, if an epitope-based patent is not allowed or invalidated, the corresponding antibody medicine (developed drug) could fall into a situation where the medicine is not under patent protection. There are countries in which examination is stricter than the United States, Europe, and Japan. Patent strategies for antibody medicines should therefore include obtaining an antibody sequence-based patent covering the antibody medicine (developed drug) in addition to an epitope-based patent (or to specify an antibody amino acid sequence, which ensures protection of the developed drug, at least in a dependent claim or the specification in an epitope-based patent application).