Collaborations between industry and academia are common in proteomics — but this doesn’t mean these relationships are always happy: There is often a precipitous chasm between the expectations of academic researchers, who are judged by their publications, and those of industry scientists, who have the investors or shareholders on their back and need to protect their intellectual property. How to make these partnerships a success was the main theme of a roundtable discussion at the recent HUPO meeting in Versailles.
What is most important, according to several participants, is a clear agreement upfront that should be signed by both parties. “It needs to be clearly defined who owns what in IP, and what the publishing schedule is,” said Hanno Langen, who heads the proteomics initiative at Roche in Basel, Switzerland. The agreement should also state a timeline, what the deliverables are, and address royalties. Equally important, Langen said, is to define what is not part of the contract.
Such an agreement, and the negotiations leading up to it, may bring up differences in the goals of the parties involved before it is too late. According to Stanley Hefta, executive director for proteomics at Bristol-Myers Squibb, a three-way collaboration between BMS, a diagnostics company, and a cancer hospital collapsed because their interests clashed — for example, the main goal of the hospital researchers was to publish their results.
Lloyd Segal, president and CEO of Caprion Pharmaceuticals, on the other hand, cited a successful collaboration his company has had with John Bergeron at McGill University since February 2002, which could serve as a model. This collaboration rests on an “umbrella agreement,” a broad set of rules, Segal said. According to this contract, all proteomics data from the collaboration is owned by Caprion, and the company has an option on all related intellectual property. In return, Caprion will pay McGill CA$1 million over five years. Bergeron serves as the single contractor, even for additional McGill investigators that may join later.
In many cases, one challenge is how to evaluate technologies or discoveries: some participants complained that academic researchers tend to overestimate the commercial value of what they bring to the table. “A target that is validated is still a long way away from a drug,” said Segal — often farther than the investigator himself believes. According to David Odelson from Invitrogen, his company handles this inherent risk in early stage research by sponsoring international scientists in academia and biotechnology to develop novel methods or reagents that Invitrogen then commercializes. These grants amount to up to $100,000 for one year. Interested researchers submit non-confidential one page pre-proposals, Odelson said.
Another problem is that academic scientists often stay out of the negotiations for collaborative agreements with industry, leaving these up to their institute’s office for technology transfer. However, Segal said he had noticed the emergence of an “academic entrepreneur” type, a senior investigator who has become sophisticated about industry collaborations. Bergeron, for example, hired an attorney to advise him on the collaboration with Caprion, which turned out to be beneficial for both sides, Segal said. Every PhD program, he suggested, should include a class on business concepts, which are “easier than scientists may think.” Educating scientists in business matters may indeed be an area where HUPO should become active, Langen suggested.
A further potentially sore point of collaborations is the timely delivery of results. While academic researchers might not always take deadlines seriously, industry wants to see bang for its buck when the time is up. “We are not money trees,” said Segal, stressing that Caprion expects scientists to account for the money they receive. “We take timelines and deliverables very seriously.”
According to Michael Dunn, professor of proteomics at the Institute of Psychiatry at King’s College London, academic researchers often share a vague fear of being exploited by their industry partners. Industry, on the other hand, is frequently afraid of academics being “loose cannons,” sharing results too early, for example. But the Institute of Psychiatry has had a positive experience with Proteome Sciences, he said. The private company set up a facility at the institute last year that performs fee-for-service work and collaborations with academic groups, as well as contract research for pharmaceutical companies.
Fundamentally, the most important factor for the success of a collaboration is establishing trust between the two parties, said Johannes Voshol from Novartis, adding that this may take some time to develop. “If there is no trust between the people on either side, it’s not going to work,” he said. Voshol is involved in managing an outside collaboration worth $4 million per year. Novartis, he said, spends about 30 percent of its research budget on outside collaborations, and dedicates one or two scientists to managing them. He found that decisions take a lot longer at universities than in a company. On the other hand, academic scientists are eager to publish their results in a timely manner, which companies do not permit until they have staked out their IP. His collaboration has solved this by stating in the agreement that if no patent has been filed within 90 days, researchers are free to publish.
Academic-industry partnerships: How can both parties win?
- Determine who owns the IP
- Discuss payments, costs, and royalties
- Set a publishing schedule
- Set a timeline to define milestones and deliverables
- Get investigators from both parties involved in the negotiations
- Define what is not part of the collaboration