Collaboration on a grand scale: negotiating big science

A research project can be a headache for an academic at the best of times, but what if you were the lead researcher of a study that encompassed more than 200 colleagues, based at multiple locations across the country? Perhaps enough responsibility to induce a full-blown migraine?

For Charles Swanton, a clinician scientist at the Francis Crick Institute and professor in personalised medicine at University College London, this is the reality. Professor Swanton is the lead researcher on the £14 million Tracking Cancer Evolution Through Therapy (Rx) (TRACERx) project, the single biggest investment in lung cancer research by Cancer Research UK.

The nine-year project, which began in July 2013, involves the collaboration of about 225 researchers and clinicians based at 19 centres across the UK. Overall, 850 patients will participate in the study, which aims to transform understanding of non-small cell lung cancer, and make an applied move towards precision medicine.

Besides obvious potential societal benefits, the academic knowledge base is also being widened: the project’s progress was marked by the recent publication of two academic papers in the New England Journal of Medicine and Nature.

With these results suggesting that the project is progressing well, how does one manage a “mega collaboration” such as TRACERx? Professor Swanton told Times Higher Education that there were things he did not anticipate when he first came up with the idea, and aspects of the project have since “snowballed”. 

“If I look back and see where we are now, and the complexity of the study, would I have been as ambitious at the very beginning?” he said. “I went into this with some naivety, thinking it would all be straightforward and possible without really thinking about the immense detail required to get it right. I’m just so fortunate to have a team behind me that is on board and incredibly hardworking, and has this vision for where we need to get to.

“[As] the chief investigator, I’m the person who stands up at meetings and talks a lot, but the brains and hard work have come from the [team]. I tend to think bigger picture. It’s all very well thinking bigger picture and where I want to get to in 10 years’ time, but it’s quite another getting to that stage.

"I tend to gloss over the detail occasionally when things get a bit too busy, whereas Mariam [Jamal-Hanjani, lead author of the NEJM paper and clinical training fellow at the UCL Cancer Institute] is very much focused on the detail. That’s what’s turned this into a very solid, plausible and important study.”

Even with a core team around him, Professor Swanton said that it was also important to keep tabs on the wider collaborators in the study. That involves a meeting every month where those involved  “sit down for an hour or two” and discuss the processes for the study – recruitment, tissue collection and processing, sample tracking and data.

Professor Swanton’s assessment is backed up by Caroline Dive, senior group leader at the CRUK Manchester Institute and professor of pharmacology at the University of Manchester, who, besides being involved with TRACERx, runs her own team of 80 people.

Professor Dive acknowledged running a mega collaboration “can be quite stressful”, but said there is a real “corporate spirit” involved in such projects.

“You have to have really good personal and communication skills just to get the ground rules of the collaboration worked out,” she said. “I still think the most important thing, when you’re assembling a mega-collaborative team, [is that] it’s very clear why each member of that team is there, and it’s a win-win for everybody: everybody has something they bring to the table and take away.

“Clearly, managing a little project with just two or three people is so much easier. Then you’ve got to deal with the outputs: being fair on authorship, making sure everyone’s signed up to the manuscripts – all of those things in big groups are always much more difficult. But they’re often more rewarding because you can do big science in big collaborations.”

Individual credit is an issue Professor Swanton also noted, and is something he takes very seriously. Professor Dive said there were “regular meetings with the TRACERx teams”, where this is discussed.

“We’ve worked really hard to make sure every TRACERx investigator nationally is rewarded with authorship – from the principal investigators on individual sites through to the tissue collectors who’ve been at the coal face managing the large quantities of tumour material,” Professor Swanton said. “There are a lot of unsung heroes out there, and we’ve done our best to identify [them]…and each one has been named on the two papers. We have 285 authors on the two papers – that’s unprecedented. I’ve never had a paper with that number of authors before, but that’s a reflection of quite how much effort was involved to get to this stage.”

Ramaswamy Govindan, a veteran of mega collaborations such as the Cancer Genome Atlas Project, and a professor at Washington University School of Medicine in St Louis, said “equal distribution of credit” was a major hurdle in these kinds of projects.

“Everyone needs a piece of the pie…and making sure everyone is appropriately credited is important,” he said. “There are some rules. You have to contribute something to get authorship. Most well-run research organisations that do multi-centre studies have clear-cut policies on [this]. That’s really good because it does not lead to disappointment or misunderstanding.”

Professor Govindan agreed that coordinating collaborators in other centres was a common hurdle for studies of TRACERx’s magnitude, but noted also the administrative barriers.

“In all multi-centre studies, a big hang-up is that getting the obligatory paperwork and the [ethical] approval process sometimes takes [a long time],” he said. “Most institutions have their own individual institutional review boards, [and] when you have a multi-centre study, having a centralised IRB is really helpful and really speeds things and moves the project along very quickly. I think we need to incorporate them more and more.”

john.elmes@timeshighereducation.com