In a recent decision concerning the patenting of antibodies (Amgen Inc. v. Sanofi), the US Federal Circuit Court raised the bar for certain antibody patents. The US patent and trademark office (USPTO) has issued a memorandum explaining the revised practice, and although the memorandum specifically refers to claims drawn to antibodies, there are already signs that the revised guidance may well impact on inventions in a wider field.

One of the requirements for patentability under US patent law is that the patent application must contain a written description of the invention and details how to make and use the invention (the ‘written description’ requirement). Previously, in the case of an invention relating to antibodies, the written description requirement could be satisfied through the disclosure of a newly characterized antigen by its structure, formula, chemical name, or physical properties. However, in Amgen Inc. v. Sanofi, the Federal Circuit Court criticised this ‘newly characterized antigen test’.

The Federal Circuit found this test to be inconsistent with an earlier decision (Ariad Pharm., Inc. v. Eli Lilly & Co 2010), and the statutory ‘quid pro quo’ of the patent system that requires an invention to be properly described in order to qualify for patent protection. The Federal Circuit held that when an antibody is claimed, an adequate written description of the antibody itself (other than by reference to its target antigen) is required.

The USPTO has subsequently issued a memorandum highlighting the withdrawal of the ‘newly characterized antigen test’ for written description of antibodies. The memo notes that some of the USPTO’s written description guidance pertaining to antibody claims is therefore outdated and advises that additional written description examination guidance is forthcoming.

The upshot is that innovators will now find it much more challenging to obtain broad protection for new antibodies. The court has indicated that for a new class of antibodies, the ‘written description’ requirement can be met through the disclosure of a “sufficient” number of representative antibodies.  It remains uncertain what a “sufficient” number would be, but it is apparent that the criteria will be very onerous. Universities and SMEs that do not have the resources to generate and characterise a large number of antibodies before applying for a patent will be hit particularly hard by these developments.

A further concern is that there are signs that some US patent examiners are interpreting the new memorandum in a manner that goes beyond claims drawn to antibodies per se. In particular, the memorandum is being cited during the examination of inventions pertaining to new diagnostic or therapeutic methods. I consider this to be an inappropriate interpretation of the memorandum and hope that the revised examination guidance will help to educate US patent examiners in this regard.

Meanwhile, it is worth bearing in mind that US practice with regard to the patenting of sequence-defined antibodies seems to remain unchanged; and that in other countries, notably Europe, it remains possible to obtain patent protection for a class of antibodies based on the identification of a new target antigen.

Nature journalist Heidi Ledford recently asked me to comment on some of these issues and her article on this topic can be accessed here.


“federal court decision has left universities and companies working to preserve intellectual property rights worth billions.”