Methodology & Guidelines
The purpose of this patent landscape analysis is to inform the reader of the major patents and players in the area of telomerase. It is intended as a source of scientific information or to facilitate risk management by uncovering patented technology that may need to be licensed or avoided in order to make and use a product.
Strategy for finding patent documents
Searches were performed using keywords (e.g., telomerase, splice variants), company names (e.g., Geron, Bayer), and inventors’ names (e.g., Shay). Both Patent Lens and PatBase (www.patbase.com) database were queried. The general approach to finding the most relevant patent documents is a scan of titles, abstracts, and lead claim when available. Generally, scanning titles for relevance is the quickest and easiest way to comb through many documents; when a title or abstract was suggestive, but inconclusive about the claimed subject matter, the lead claim was reviewed. In this way, a manageable list of about 100 documents was obtained. A second review of this list entailed a review of the independent claims. Ultimately, a final list of documents was selected.
Decisions were made to exclude certain documents. In particular, if the only document in the patent family were a Japanese patent or patent application, the document was omitted from analysis. The only substantive information from these documents was a translation of the abstract, which may not accurately reflect the claims. In addition, for areas that contained U.S. patents, U.S. applications were not analyzed. Updates of this report will consider any that have been granted in the interim.
Limitations of patent searches
Two main types of limitations may affect the outcome of a search: those that are inherent in the data and those resulting from the search process. While every effort is made to minimize the limitations and their effects, inevitably there will be some. These limitations and their effect are discussed in this section to promote appropriate and informed reliance on the patent data and analyses.
Data are not error free. Common types of data problems include misspellings, alternate spellings especially of names and assignees, inconsistency of patent examiners in classifying patent subject matter, translation errors (especially for Japanese documents, but also for European patent documents). Some of these errors can be overcome by using wildcards in formulating the search or broadening the scope of the search; however, other errors inherent in the data cannot be neutralized.
Another source of possible problems arises from the INPADOC data, which includes patent information of more than 70 countries. Because countries use a variety of different update schedules (two weeks to one year) and report different types of information (e.g., not all report legal status), it is not always possible to verify filing or legal status of a patent application in national patent offices.
Furthermore, the search process may engender additional consequences. Although using a combination of key term and classification criteria will usually capture all the key patents and applications in a field, the downside is that too many documents to realistically further screen may be found. When there is a substantial number of documents recovered (more than a few hundred), or the primary document is not in English, the only practical means of identifying potential key documents is on the basis of the title and abstract, which may or may not represent an accurate picture of the claimed subject matter.
As well, patents are dynamic. On a steady basis, new patent applications are filed and published, new patents are granted, patents are abandoned, and the law changes too. Therefore, this report is but a snapshot of the landscape as it appeared in September 2006.
Claim construction rules
The interpretation of a claim is primarily based on the plain language of the claim and on definitions, explicit or implicit, in the text of the patent document. Of course, claims in patent applications have not been examined and may well be different when granted (if granted at all). For granted claims, a more precise interpretation requires a reading of the prosecution history – the written record of the examination process between a patent office and the patent owner – for instances of the patent owner stating what the claims mean or giving up claim scope.
Patent claims are intended to provide notice to others of the scope of the protected invention. Unfortunately, more often than not, reality falls short of the ideal. Part of the challenge, when reading a claim, is to set aside your own biases about what words mean and question the definition of each term. While technical terms beg definition and thus often become the subject of dispute, common words may also be contested. Recently, patent claim construction in litigation has sometimes centered on disputes regarding definitions of words as “about” and “adjacent”, among others.
Claims often use terms, especially technical terms, that require a bit of detective work to understand their meaning. The steps of investigation include identifying terms that need defining (generally, words other than common English words), looking in the text of the patent document for an explicit definition or use of the term, look in the prosecution history for definition or use of the term, and consulting a dictionary or expert in the field, if available. Pursuit of some of these steps is not within the scope of this report; the order that the steps are performed and the weight given to each is currently the subject of hotly contested litigation, but for our purposes starting with the plain language of the claim and incorporating definitions from the text of the patent will provide a useful working interpretation.
While reading claims, a very important rule of claim construction is useful to keep in mind: when the term “comprising” is used, a device that has more elements than the claimed device falls within the scope of the claim. Thus, if the claim recites “an antibody comprising A, B, and C” and there is an antibody made up of A, B, C, and D, the second antibody falls within the claim. Conversely, an antibody made up of A and B does not fall within the claim.
Other useful rules:
- the term “a” or “an” means one or more;
- a patentee can write her own definitions of terms;
- claims are interpreted by the current law, not the law at the time of grant.
Note that in addition to issued patents, patent applications are included in this report. Sometimes, the claims of patent applications are written so broadly that it is not readily apparent what claim language is likely to be granted, if there is a grant.
Content of survey
In the following section of this report, an overview of the patent landscape is presented. Attention is drawn to the key patents – those that appear to have the broadest claims. For a number of reasons, the report analyses mainly patents sought or granted in the United States: the largest key player, Geron, is based in the United States, U.S. patents are readily obtained, the law regarding claim interpretation is most well developed. In contrast, because patents granted by the European Patent Office are interpreted according to each country’s laws, the body of patent law isn’t homogeneous.
The documents discussed in this report are a mix of patent and patent applications. For patent applications, caution is warranted when evaluating the claim scope because the claims have not been examined and may differ if and when granted.
The format of the survey generally contains a summary of the patent description, a listing of the most relevant independent claim(s) and discussion of the claim meaning.