"Pharma employs some of the smartest people on the planet, but then occasionally makes the mistake of telling them what to do. You cannot tell bright people what to do and then expect them to be passionate, creative and have ownership of the project. You cannot be time-line driven, you cannot be commercial-value driven you have to be science and data driven," says Dr Chas Bountra, chief scientist, Structural Genomics Consortium (SGC), University of Oxford.1 Just another ivory-tower academic sounding off? Not quite. Dr Bountra spent 20 years in pharmaceutical drug discovery at GlaxoSmithKline (GSK), taking three molecules to Phase III and one molecule to market, before being approached to lead an elite team of 75 scientists at SGC Oxford in January 2008.
Founded in 2003 as a not-for-profit organisation by Big Pharma (GSK, Merck, Novartis), the Wellcome Trust, and government and charitable organisations in Canada and Sweden, SGC, with an annual budget of $30m, specialises in structural biology with the group in Oxford having a disease focus in oncology solving the 3D-structure of human soluble and membrane proteins at a phenomenal rate of five per month. Dr Bountra puts this into perspective: “Of all the kinase structures that have been solved in the last three years globally, Stefan [Dr Stefan Knapp], one of our Principal Investigators, has solved the structures of half of them." Moreover, SGC Oxford is doing this more cost-effectively than other groups in academia or pharma: "We solve them for about £100000 because we have focus, absolute commitment, and are obsessed about delivery and efficiency we have built up almost an industrial process," he says.
Data for all
All data is placed in the public domain; no attempt is made to secure any intellectual property on protein structures or the chemical probes being used to identify potential targets for new therapies. In contrast to the pharmaceutical industry's model of securing intellectual property rights as early as possible, SGC's model of 'pre-competitive' chemical biology is changing the landscape of early drug discovery for pharma by freely providing druggable targets, and well characterised chemical probes, and accessing medicinal chemistry to do lead optimisation themselves. This would not be possible without support from Big Pharma, however: "GSK is helping us to change the paradigm. Senior individuals have agreed to devote eight chemists for the next four years at cost to them and it's quite a paradigm shift here, since these eight chemists are going to make compounds which we are going to put into the public domain," continues Dr Bountra.
This arrangement suits Big Pharma funders, who nominate protein targets but who do not get prior access or preferential data rights, and the public backers, who acknowledge that it is better for society and patients that fewer parallel clinical trials are conducted on first molecules for novel targets that rarely, if ever, become drugs. Dr Bountra does expect that once chemical probes have been used to validate the target and the data is published, pharma will act: "If we publish the data, the protocol, the details of the molecule, efficacy achieved and side-effects, pharma will jump on that target and then they can go in parallel, do their chemistry, run screens, take out IP and hopefully get a drug."
The approach at SGC follows in the footsteps of the pre-competitive academic consortia established in the late 1990s such as the Dundee Kinase Consortium, University of Dundee, UK, (1998), and the SNP Consortium, Cold Spring Harbour, US (1999).
However, a number of different industry–academic vehicles exist. A current paper in Drug Discovery Today by Cathy Tralau-Stewart et al at Imperial College London shows that industry is collaborating with academia through proof-of-concept funds, corporate mini-labs, industry-state funded research centres, and sponsored research.2
The more entrepreneurially-minded universities have set up their own drug discovery units to translate projects from the bench to the clinic, yet Dr Tralau-Stewart concedes that the well-documented pressures of publication versus patent remain. As Dr Bountra says, "For us to maintain academic credibility here in Oxford, we have to produce publications. If we are going to attract more money, we have to produce them, and it is a way of ensuring our science is top-notch."
With recent citations in Nature, Cell, and PNAS, it is no wonder that the Wellcome Trust invested a further £4.1m in SGC Oxford, specifically to deliver characterised chemical probes for proteins involved in epigenetic signalling. Along with the money comes in-kind contribution worth £10m of high-throughput screening over the next four years, supplied by the NIH in Washington and the chemistry capabilities at GSK. Despite success in structural biology and chemistry, Dr Bountra readily concedes that SGC cannot do everything, and explains why they manage over 100 scientific collaborations: "The great thing about working here is that when you don't take out IP, everyone wants to collaborate with you. Imagine I could go to Imperial College and say 'I have the 3D structure for this target, or I have this probe, would you put it into your assay?'. We'll get a joint publication out of it this couldn't happen so easily in pharma."
One of the drivers of industry–academic partnerships, according to Dr Tralau-Stewart's paper, is that academia is well-placed to consider and anticipate potential hypotheses for project attrition, citing the search for the universal cure-all blockbuster with broad market penetration, and misleading animal disease models. Dr Bountra's experience with animal models in industry suggests that one cannot automatically assume a molecule will work in the clinic because it worked in an animal model: "Ultimately, target validation happens in the clinic, not in animal models."
Industry may also learn from the creative and dedicated culture of excellence that can be fostered in academia under the right conditions, as Dr Bountra observes: "These people will not accept failure they have ownership, passion and commitment; for example Frankie [Dr Frank von Delft] our crystallographer, routinely works 72-hour stints literally day and night. Unless you say 'I really want to do it, I care about this project, and I want to make it a success,' its not going to happen. These guys have it."
Having the right culture, focus, talent and metrics is a rarity in both industry and academia, and Dr Bountra thoughtfully concludes: "I think we sit somewhere between industry and academia."
1. Author interview with Bountra, C., 12 December 2008, Oxford, UK.
2. Tralau-Stewart, C. et al, "Drug discovery: new models for industry-academic partnerships," Drug Discovery Today 14, 95–101 (2009).