The patent system exists to advance innovation by encouraging inventors to share their ideas publicly. In return for revealing how their invention works, inventors receive patent protection. Patents should therefore be comprehensible to professionals working in the relevant field, not just patent attorneys. However, unlike peer-reviewed journal articles, patent applicants have no incentive to draft engaging or easy to navigate patent applications, beyond what is needed to satisfy the formal requirements of the patent office. The result is patent documents that appear rather impenetrable on first glance. However, there are a few tricks of the trade that make reading and extracting the key information from patents much more straightforward. This particularly applies to the biotech field, in which the data provided in patent applications can be a mine of useful information.
Let us imagine you are a scientist who has just had an exciting idea for a project. You are interested in whether someone else has already patented the idea. A collaborator may have already forwarded you a patent that seems relevant. How do you check whether the patent is a problem?
Where to search
Google Patents is a free and effective tool for a preliminary patent search. Advanced search gives the best results. In the Advanced search options, there are the options to do a keyword search, filter by patent office and status, as well as searching for particular inventors or assignees. It is also possible to limit the keyword search to parts of the patent document. “CL=” will search just the claims.
Is it granted?
The first thing to check when you have found a patent document is whether it is granted. A patent application cannot be enforced until it is granted. The scope of the patent will also not be clear until it is granted. The information box on the right hand side of a page in Google Patents provides the status. Additionally, when looking at a particular document, the official number will also tell you whether the patent is granted. If the number ends with an A and then a number (e.g. A1) the document is a patent application. If the document ends with a B and then a number (e.g. B2), the document is a granted patent.
Where is it granted?
It also matters in which country the patent is granted. Patents are country specific. A US patent can only be enforced against activities in the US, whilst a Japanese patent can only be enforced against activities in Japan. The Google Patents record will indicate which country the patent is in. The first two letters of a patent number are also the country code.
Patent application numbers beginning with a WO are international (PCT) patent applications. Importantly, it is not possible to get an international patent as such. PCT applications have to be examined individually in the separate countries and jurisdictions, and are then granted as US, JP etc. patents.
Has it expired or lapsed?
If the patent is granted, the next question to ask is whether it has expired, abandoned or allowed to lapse. The expiry date for a patent will also be listed in the status box for a patent on Google Patents. An expired patent cannot be enforced against new activity.
The primacy of the claims
The claims are the most important part of the patent. Unlike for academic articles, reading the abstract of a patent is fairly pointless, as it will not give very much information about the invention. The introduction sections of biotech patents are also usually very generic. The claims are the starting point and the basis for assessing the patentability of an invention (G1/24). Therefore, once you have established whether you have a granted patent and whether it has expired or lapsed, you should next read the claims.
The claims define the scope of the protection conferred by the patent. A claim that does not depend on any other claim (e.g. does not refer to “the invention according to claim 1-3), is referred to as an independent claim. The other claims, which refer to previous claims, are the dependent claims. The independent claims define the invention in its broadest form. The broadest claim will usually be the first claim. However, there may also be broad independent claims further down the list.
To directly infringe a patent claim, a product must have all of the features specified by the claim (unless the claim indicates that the feature is optional). The language “comprising” used in patent claim means the claimed invention must include but is not limited to the specified features. For example, a claim may specify a molecule comprising a VH and VL region. A molecule comprising just a VH region and not a VL region would not directly infringe this claim. A molecule comprising a VH, VL and Fc region would infringe the claim. By contrast, the language “consisting” means that the following features, and only those features, are included.
The relevance of the description: The examples are key
In biotech, the description is always highly relevant to the claims (even before G1/24!).To satisfy the requirements for patentability, biotech inventions have to be supported by data. The data is found in the examples section of the description. The level of data required to support a biotech invention varies depending on how broad the claimed invention is and how technically unpredictable the field. The broader the invention, the more data that is usually required. Similarly, inventions in highly unpredictable fields will require more data.
If you are reading a patent application that has not yet been granted, the correspondence between the data and the claims can also help you predict the likely scope of a patent once it is granted. Patent offices will in many cases force the patent applicant to limit the claims in scope to correspond to the level of evidence provided by the data. For example, a patent application claiming Drug X to treat autoimmune disease, may have to be limited to a claim for Drug X to treat rheumatoid arthritis if the only data in the Examples relates to the treatment of rheumatoid arthritis.
The examples are also where you may get hints as to the lead or likely commercial embodiment of the invention. A patent application may, for example, list hundreds of molecules but only provide efficacy data for one molecule. It is likely that the molecule with the data is the lead.
However, it is also worth bearing in mind that the level of data required to support a patent is often lower than might be expected. The data needed to support a claim for a new therapy, for example, is usually far lower than is required by regulatory authorities or even peer review publication. The Boards of Appeal of the EPO have upheld patents, for example, relating to the treatment of hard to treat neurological diseases, in which the only data provided was quantitative pre-clinical data without evidence of statistical significance. See also the infamous “fish juice” case, in which a patent for the broad use of fish oils and (any) juice to treat cancer, was also upheld. The key principle, however, is that there must be some form of evidence that renders the claimed invention plausible.
Sequence information
Biotech patents will often contain sequence information. Any sequence included in a patent (whether claimed or not) must be provided in the format of a sequence listing so that the patent office can search for the sequences in the prior art. If the claims of the patent refer to e.g. SEQ ID NO: 1, the best way to find the actual sequence of SEQ ID NO: 1 is therefore to refer to the sequence listing. Unfortunately, the sequence listing is not provided on Google Patents. If there is an international (WO) application in the patent family, the best strategy to find the sequence information is to copy the application number (e.g. WO 2009/XXXXXX), and search for this case in the WIPO database. For newer patent applications, the WIPO record will include a sequence listing tab providing the sequence listing (see for example, WO2025086418). For older applications, the sequence listing will be provided as a XML/TXT file in the documents tab.
The relevance of the detailed description
It is only now, after we have read the claims, examples and sequences, that we usually turn to the rest of the description. The main purpose of the detailed description is to provide basis for the claims and any future amendments to the claims that the applicant might wish to make. The detailed description must therefore contain, as far as is possible, all the back-up positions that the patentee may need during prosecution. The European Patent Office has a very strict rule against adding any subject matter to the patent application after the filing date. This section of the description is consequently usually full of highly repetitive text (wherein the compound may be X, wherein the compound may be Y etc.) and long numeric lists (e.g. 1 mg, 2 mg, 3 mg, or 4 mg etc.). The description can therefore be very cumbersome to read.
Thankfully, unless you are conducting an in-depth validity analysis, this section of the description can usually be skimmed over. The detailed description becomes relevant for validity and FTO analysis as to whether there is adequate basis in the description for the granted claims and/or any amendments that the patentee may wish to make to cover competitive products.
If you are particularly keen, you may also wish to compare the sequences provided in the detailed description to that of the sequence listing, and the public sequences of the commercial/lead product (e.g. as provided in the recommended INN). Unfortunately, sequence errors in patent applications are extraordinarily common.
G1/24: Definitions and context
The last few days have seen a splurge of articles and social media posts attempting to explain the impact of the recent Enlarged Board of Appeal (EBA) decision in G1/24. This decision related to how the claims of a patent should be interpreted, and particularly whether or not the description should be consulted. The EBA in G1/24 stated that the description should always be consulted in order to interpret the claims. In biotech, given the need for the description to contain data supporting the claimed invention, it was always necessary to consult the Examples and sequence information provided in the description when evaluating the scope and validity of a claim. G1/24 did not change this.
So what is the relevance of G1/24 to biotech patents? In addition to the Examples and basis for the claims, the description of a patent may also contain definitions. Given that G1/24 tells us that the description should always be referred to when interpreting the claims, any definitions provided in the description will need to be consulted. As such, what appears to be a very clear limitation in a claim may be interpreted more narrowly or broadly than usual because of a definition in the description.
Whilst a patent may, for example, claim “an antibody”, there is a spectrum of possible definitions for the term “antibody” that the patentee might use in the description. An antibody may be defined very narrowly, e.g. limited to monoclonal antibodies, having VH/VL and Fc regions. An antibody may alternatively be defined more broadly as including other forms of binding molecule, e.g. a scFv lacking an Fc region. Alternatively, an antibody may be completely redefined to mean a chimeric antigen receptor. The definition of the term “antibody” is also likely to be influenced by the rest of the description, and what data is provided e.g. whether the data relates exclusively to scFvs or CARs. It is at this point that interpretation of the scope of protection of the patent becomes more complex. The description is consulted as a whole, and therefore the scope of the claims may be influenced by the various definitions, examples and choices of terminology provided in the description.
Final thoughts
Navigating the dense landscape of patent documents can initially appear daunting. However, by understanding the structure and key sections of a biotech patent, it is possible to find a wealth of valuable information from the claims, examples and sequence information. The decision of the Enlarged Board of Appeal in G1/24 is a reminder, however, that determining the eventual scope of protection of the claims can be difficult to predict, especially in biotech where the disclosures provided in the examples and sequence information are inextricably linked to the claims. Therefore, whilst reading the claims should always be your starting point when reviewing a patent, a comprehensive analysis must also include a detailed review of the description.
Further reading
- When does pre-clinical data plausibly support a therapeutic effect? (T 966/18)
- Fishy evidence: Do we have the right sufficiency standard for therapeutic use? (T 1057/22)
- Freedom to operate versus patentability in biotech: What the difference is and why it matters
This post is based on a previous article on IPKat.