This article has been updated from a previous version to clarify data from the study.
About half of all National Cancer Institute technology licenses are executed within three years of the technology's initial patent application date; while almost all licenses include technologies within eight years of the application date, according to the results of a recent NCI study on the influence of patent term on biomedical technology licensing.
The results suggest that tech-transfer offices at academic, government, and non-profit research institutions seeking to out-license biomedical innovations should continue maintaining unlicensed patents older than three years, but may be wasting their time and money if a patent is more than eight years old, according to the study's authors.
"Our data suggests that it is unlikely that patents which are unlicensed more than eight years after priority patent application filing will ever be licensed," the study's lead author Keith Bupp, of the marketing and metrics unit of the NCI Technology Transfer Center, wrote in an e-mail to BTW this week.
"Maintaining these older unlicensed patents by paying maintenance fees – and possibly continuing patent prosecution costs – is therefore probably not usually warranted unless there is significant ongoing development by the NCI," Bupp added.
Bupp presented results of the research during a poster session at the Association of University Technology Managers' annual conference, held last week in Orlando, Fla.
The study is part of an ongoing effort by NCI's Technology Transfer Center to analyze "the factors that determine whether or not a patented technology will be licensed," Bupp said.
"Ideally, funds would only be used to patent those technologies that will be licensed," he said. "However, it is frequently difficult to predict in advance which technologies will become licensed."
The ability to make these predictions is especially crucial in the biomedical and pharmaceutical sectors, as evidenced by NCI's data.
The poster summarized a portion of existing commercial products that include NCI-licensed technologies, such as the human papillomavirus vaccines Gardasil and Cervarix; a chemotherapeutic treatment for B-cell chronic lymphocytic leukemia; layered peptide arrays for analyzing cancer fingerprints; monoclonal antibodies to P450 enzymes for drug toxicity studies; PathVysion HER-2 DNA breast cancer diagnostic; PixCell laser capture microdissection system; multiple myeloma treatment Velcade; and Zevalin, a mAb for treating non-Hodgkin's B-cell lymphoma in conjunction with Rituxan.
According to the NCI study, which has not been published, the cost of bringing a single drug to market can reach $1 billion and take an average of seven years from the start of clinical trials.
"In order to recoup this investment, companies require temporary market protection afforded to them by patents," Bupp and colleagues wrote in their poster. "When considering whether to license and develop technologies, patent life is therefore an important factor for industry."
In order to understand how a patent's age correlates with licensing, Bupp, along with NCI colleagues Bonnie Chamberlain and Karen Maurey, examined all the technologies licensed from the NCI whose patents or patent applications had priority dates on or after June 8, 1995.
"Patent life for patents with priority dates on or after June 8, 1995 in the US is determined by the date of application filing," Bupp explained. "This was not the case before June 8, 1995. It was thus simpler to normalize the timeline data using [that] date of filing as a cutoff point.
[ pagebreak ]
"Also, by using this date, the information presented reflects more current practices in licensing of NCI technologies than would have been the case if licenses of older patents were included," Bupp added.
They defined time of licensing as the date a new license agreement was executed, and considered the duration between the date the patent priority application was filed and the date the license was executed. In cases where multiple technologies were included in an agreement, the researchers used the technology that had the shortest duration between application and first license.
Overall, the data included 261 licenses and 234 inventions. The numbers of licenses by license type included three that were "biological material-commercial;" seven that were for "biological material-internal use;" 57 that were for "commercial evaluation;" 159 that were "patent-commercial;" 28 that were for "patent-internal use;" one that was defined as "settlement-infringement," five that were defined as "settlement-interference;" and one that was "settlement-litigation."
The study revealed that approximately 40 percent of licenses were executed within two years of a technology's patent priority date; and approximately 50 percent of all licenses were executed within three years of that date.
"Thus, in order to maximize the potential licensing and commercialization of inventions, it may be necessary for a [tech-transfer office] to consider maintaining patent protection of unlicensed applications beyond the statutory deadline for national phase filings" — typically two and a half years after the filing date — the researchers wrote in their poster.
"If applications are unlicensed [after the statutory deadline], this means that NCI would have to pay for these applications rather than rely on royalties from a licensee," Bupp told BTW in his e-mail.
Conversely, about 93 percent of all licenses were executed within eight years of the patent priority date. "Thus, maintenance of patent protection for longer periods may not yield additional licenses," the researchers concluded in their poster.
The researchers wrote that the sharp decline in licensing activity noted after the eighth year is likely due to a number of other factors, including the limited remaining patent term, which they said "significantly reduces the incentive for the licensee."
They add that their results "provide a useful time frame when patent prosecution and maintenance can reasonably be considered, particularly for early-stage FDA-regulated biomedical technologies such as those developed at NCI."
Bupp and colleagues also note that that the state of the technology, continued development by the inventor, and market factors all play an "important" role in determining whether a technology is licensed.
Another factor could be how aggressively a licensing institution markets its technology: If a technology has gone unlicensed for several years, it stands to reason that any tech-transfer organization, including NCI, would not be as likely to continue marketing that invention to potential licensees.
"There are multiple factors that will determine whether or not a technology is licensed at a given time, including its state of development, novelty, and other considerations," Bupp said. He added that NCI technologies are advertised through the Federal Register and through a variety of websites, e-mail lists, and conferences, though "we have not examined the aggregate effects of our marketing efforts on licensing."
Bupp also said that the researchers do not know if their data is endemic only to the biomedical industry, or whether they would hold true for industries such as engineering and information technology.
He said the team next plans to examine NCI's patent portfolio to see if different technological sub-categories — for instance vaccines, chemotherapeutics, or research tools — would demonstrate different durations between patenting and licensing.
"We are interested in analyzing our patent portfolio for licensing trends with respect to these and other characteristics of NCI technologies," Bupp said.