Donate to Science & Enterprise

S&E on Mastodon

S&E on LinkedIn

S&E on Flipboard

Please share Science & Enterprise

Analysis Uncovers Biotech Commercialization Bottlenecks

Jerry Thursby, Matthew Higgins and Marie Thursby

L-R: Jerry Thursby, Matthew Higgins and Marie Thursby (Georgia Institute of Technology)

21 August 2014. The path from scientific discovery in an academic lab to the marketplace is rarely a straightforward process, with researchers and entrepreneurs often facing detours and delays keeping new biomedical technologies in limbo for years at a time. Those are the conclusions of faculty at Georgia Institute of Technology’s business school that studied the fate of hundreds of patented inventions in university labs in an analysis appearing in this week’s issue of the journal Science Translational Medicine (paid subscription required).

Jerry Thursby, Matthew Higgins, and Marie Thrusby — professors of entrepreneurship and management at Georgia Tech’s Scheller College of Business — reviewed the disposition of 835 patents for biomedical discoveries from university research labs. The analysts looked particularly at the licensing processes first between the university and the initial licensee, often a small biotechnology company, and later between the biotech company and a larger pharmaceutical enterprise with the resources to carry out lengthy and expensive clinical trials.

The review by Thursby, Higgins, and Thursby revealed the process for translating discoveries into marketable technologies is rarely quick, smooth, or straightforward. The 835 patents resulted in 342 licenses mainly with biotech businesses, a process taking on average 5.5 years. Nearly all (92%) of these first licenses occur in the early discovery stages, when the initial molecules are identified or defined.

Of those initial licenses, 27 percent were subsequently licensed for clinical development, a process taking another 3.5 years on average. “The timeline for commercialization is much longer than most people think,” says Jerry Thursby in a university statement. “There is so much turmoil and churn within the process.”

That 27 percent slice may seem small, but the authors show that biotechs often uncover other disease targets for the initial discovery between the first and second licenses. In nearly half (44%) of the cases, none of the first-license disease targets were part of the second licenses for clinical development. Many of the the remaining subsequent licenses added targets (28%) to the initial set of disorders, or both added to and reduced the list of targets (9%).

When targets are replaced or added, however, the discovery and development process often starts over, with the clock resetting as well, and more time added that process. The researchers call this starting-over syndrome, “bench-to-bench”.

The authors cite some highly successful drugs, both in terms of clinical benefit and finances, that were developed as added targets from the original. Pfizer’s drug Viagra, for example, was discovered as a treatment for erectile dysfunction during trials as a therapy for cardiovascular conditions. Propecia by Merck, was first approved as a treatment for enlarged prostate and later discovered to treat male-pattern baldness.

One of the reasons new disease targets are not pursued, however, is the narrow research focus of academic labs and small biotech companies that often lack the resources to do the extra work to find more uses for their discoveries. Another factor, say the authors, is the mixed incentives, at least among academic scientists, to take part in translational research.

The Georgia Tech team investigated a larger pool of 948 university licenses and found a solid majority (62 percent) of the agreements had some incentives for faculty to take part in development of drugs from their discoveries, such as funding for the researcher’s lab or a percentage of the milestone and royalty payments. However, nearly the same percentage (60%) of the licenses gave the licensee companies a right to delay publication of findings related to the licensed technology, for about 3 months on average, but in some cases as long as 18 months.

The researchers recommend a set of policies and practices to reduce the delays and bottlenecks. Among the suggestions is a database of biomedical discoveries in preclinical development, modeled after and linked to This open database would list biomedical patents and licenses, cross-referenced with once the discoveries reached the human trials stage. Universities and biotechs would be required to index their discoveries in this database, but the program would also include rewards such as extended patent protection for new targets from earlier discoveries, particularly after drugs reach the generic stage.

Read more:

*     *     *

Comments are closed.