Help scientists beat cancer (while having fun)! By Amanda Coutts, November 10th 2014
Dr Anne Kiltie is a scientist who runs a research group at the University of Oxford in the Department of Oncology. Anne has recently been involved in the development of a new Citizen Science project: Reverse The Odds which is a unique mobile game commissioned by Channel 4 for Stand Up To Cancer where users will analyse images of real cancer cells to aid researchers extract vital information to help beat cancer sooner.
We spoke to Anne about her work and how she got involved in this project.
AC: Tell us about your current research.
AK: I’m investigating what happens inside a cell when its DNA is damaged – specifically how the cell is alerted to the damage and the repair process. The research covers two different subject areas within this field, both relating to muscle-invasive bladder cancer (MIBC).
By understanding how cells respond to DNA damage, we can hopefully develop more sophisticated treatments.
I’m particularly interested in a type of DNA repair process called microhomology mediated end joining (MMEJ). Because there are several ways to repair DNA within cells, the cell in question must choose the right one to fix the break. MMEJ is a particularly error prone repair mechanism and can cause increased levels of mutations – leading to the rearrangements in DNA that are a hallmark of cancer. Because of this, MMEJ is usually a last-ditch repair mechanism for cells with DNA damage – but in certain types of MIBC, MMEJ is the preferred repair process.
I’m also investigating a protein called MRE11. It acts as roadside assistance for cells. When a cell’s DNA is damaged, MRE11 arrives at the scene with a team of other molecules to assess the situation. If the DNA can be easily repaired, it’s quickly mended and the cell continues to divide as normal. If the damage is more serious, the cell is taken “off the road” while the damage is repaired. And if the damage is beyond repair, the cell is “written off” and destroyed – preventing the accumulation of mutations in our cells. We’ve already shown that the presence or absence of MRE11 in MIBC can be used to predict how well a patient will respond to radiotherapy.
Now we want to increase our understanding of the underlying biology and use our findings to improve patient outcomes. We can do this in two ways. Firstly, by exploiting the mechanisms to develop new treatments to make cancer cells more sensitive to radiation. This means we could add drugs to radiotherapy that kill more tumour cells, while sparing our healthy tissue. Secondly, by discovering and developing relevant biomarkers to predict patient outcomes and improve treatment choice.
AC: What are the big outstanding questions in the field?
AK: A big objective in the field at the moment is to enable patients to be able to make an informed choice about their treatment options.
Patients with muscle-invasive bladder cancer are treated by either surgical removal of their bladder (cystectomy) or radiotherapy-based bladder preservation. Both of these options may also be supplemented with chemotherapy. Unfortunately, there is no clear evidence to tell us which approach is better, so patients have to decide between the two.
We want to find molecular markers – such as MRE11 – that can indicate how well a particular cancer will respond to the different treatment options. Testing a tumour for these markers before definitive treatment would enable us to provide patients with the necessary information to make a much more informed choice.
As each patient would be increasing their chances of tumour control, this should result in increased cure rates overall.
We’re testing a range of molecular markers, not just those relevant to DNA damage and repair. In doing this, we should hopefully identify markers which can predict who will do better with radiotherapy-based treatments or those who would benefit more from a surgical approach. It may well turn out that a combination of markers is required rather than just one.
By outsourcing data analysis to the public with Reverse The Odds, we’re able to speed up the analysis of 18 different biomarkers in MIBC samples. Without it we could only have looked at 10 of these. Not only are citizen scientists speeding up the research – they’re also enabling us to do a lot more with what we have. Hopefully this will lead to the discovery of new prognostic markers.
AC: So you have recently been involved in the development of a Citizen Science project can you tell us a little more about this and how you got involved?
AK: My lab is producing colossal amounts of data and getting through this data can take years – especially as much of it can only be analysed by the human eye. That’s why we were keen to explore innovative ways that could help us get through this data sooner.
Cancer Research UK approached me earlier this year for my opinion on current and future demand for pathology analysis in research. A couple of months later, I was invited to be involved in a citizen science game that was being developed in collaboration with Maverick Television’s Multiplatform team, Chunk and Cancer Research UK. The game would be scoring our samples and I was enthusiastic to be a part of this pioneering initiative.
The game – Reverse The Odds – is based around engaging puzzle challenges that can be played on mobiles and tablets. Players are asked a few quick questions about images of cells – which are from patient samples of bladder cancer. Each image is looked at by lots of people so their answers can be averaged out to give us a highly accurate idea of what’s happening in each sample.
The puzzle element gets harder as the game progresses and the player is incentivised to do more scoring by getting extra ‘potions’ to help them advance. The combination of the engaging game element and the scoring element means that people can have fun while at the same time aiding real cancer research.
One of my responsibilities was to work on the tutorial questions to ensure citizen scientists understood the questions and could give accurate answers. Meanwhile, the game developers were busy their end, and I got to see the prototype game during a very exciting and productive face-to-face meeting of the project team in London. Having ironed out a few technical issues, things moved very quickly and the game was launched in October 2014.
AC: How many samples are in the Reverse The Odds and will you add to this on a continual basis?
AK: I have over 21,000 samples in the game that need to be scored by citizen scientists This will encompass the classification of 18 different biomarkers in samples of muscle invasive-bladder cancer. Each sample is then split into 36 parts which is looked at by multiple players
Colleagues from Southampton also have images from lung cancer which are being scored. One of the strengths of Reverse The Odds is that other researchers will be able to add their datasets in the future.
AC: Besides additional manpower are there other advantages to using a citizen science project to aid in data collection?
AK: We would never have been able to do this whole project on our own, so the additional manpower provided by citizen scientists is key. The Citizen Science team at Cancer Research UK work with the online citizen science developer and platform, Zooniverse, to do all the processing of the data. This means we are given a result that can be directly compared with our results in the lab. Now we only have to score 10 per cent of the images for comparison with the scores of the citizen scientists – this saves us a lot of time and the results we’ve seen so far are very encouraging.
In addition, having multiple responses on each image should make the results more objective and robust than only taking the opinions of the small number of people who would normally look at these samples.
The final advantage is that Reverse The Odds has provided a channel with which to engage the public in our research and enable them to take a direct role in beating cancer sooner.
AC: How do you get the data from the app and how will this data help your research?
AK: We will be sent data periodically from Zooniverse. We will first compare the results with the 10 per cent of images we have scored to check for consistency. Then we will see if there is any link between the biomarkers and patient outcome data. Any interesting results will be validated in further studies – hopefully meaning we will find a panel of biomarkers that can be used in future to predict patient outcomes.
AC: Are you planning on getting involved in any more projects like this?
AK: The whole experience has been very positive and rewarding. If we can show that citizen scientists are at least as accurate as researchers, and we’re able to get through the data faster, we’ll want to look at more ways to get the public involved.
AC: Can you see this type of Citizen Science project being of benefit to other researchers with similar questions/research problems?
AK: Yes, absolutely. Reverse The Odds has allowed the lab to expand our research to study a much wider array of molecular markers in MIBC samples. I would not have had the time or resources to do this otherwise. It’s an innovative tool to broaden and accelerate research for many labs.
AC: Did you play the Reverse The Odds a lot to test during development?
AK: I didn’t personally play the game but Channel 4 and Cancer Research UK employees ‘road-tested’ it, and we had about 80 external testers analyse slides. We compared their results with our ‘expert’ results and found the comparison to be very encouraging.
Since the release of Reverse The Odds, the members of my lab have all been playing Reverse The Odds and their feedback has been extremely positive.
If you'd like to become a citizen scientist you can download Reverse The Odds below!