Parkinson's - the Big Data Opportunity

This week is my fourth anniversary but it’s not one I will be celebrating. On Wednesday January 2nd 2019 a doctor at St Mary’s Hospital in Paddington told me I was exhibiting clear signs of what she called “Parkinsonism”. It would be confirmed later by a scan but it was not a big shock to me - after all, a neurologist had written to the BBC a few months earlier to warn that I might have Parkinson’s Disease after seeing my hand shaking in a live television broadcast.

And I didn’t have time to brood - three days later I was on a plane to Las Vegas for an exhausting and manic week covering the world’s biggest gadget show. Running around the giant noisy halls of CES trying to find the latest kooky robot, and then heading to Phoenix, Arizona, to make a radio documentary about driverless cars was a useful distraction.

But when I got home I started work on a mission which has occupied much of my time over the last four years - trying to understand Parkinson’s, its treatment and the research that might lead to a cure. I have applied to take part in clinical trials, experimented with various gadgets, visited the Parkinson’s Brain Bank and spoken to some of the leading researchers.

One thing that has become clear to me is that understanding just how the disease arrives in and progresses through the brain is a key challenge and one where researchers need a lot more data. So I was very interested in a project to build the biggest database in the UK - possibly in the world - tracking people with Parkinson’s over many years .

It goes by the rather uninspiring name of PC3 which, as the man leading the project Professor Alastair Noyce explained, stands for Parkinson's Clinical Cohorts Collaborative. What it involves is bringing together as many as a dozen existing projects which have recruited cohorts of people signing up to be studied, linking up their medical records and creating one vast resource for researchers into Parkinson’s.

Professor Noyce is a neurologist treating patients but spends most of his time as an academic at Queen Mary University of London leading a big team of Parkinson’s researchers. There, he runs a couple of the studies which could be part of PC3, the East London PD project which looks at Parkinson’s in diverse underserved populations and PREDICT-PD, a UK wide study to try to identify who's at risk of developing the disease.

It was in going through the difficult process of getting approval from the NHS to link the data collected from participants in the PREDICT-PD study with their health records that the idea for PC3 struck him:

“It's incredibly laborious and incredibly expensive to try and achieve records linkage between data that's collected in a research setting and then the NHS health records. And actually, we've got 10 or 12 high quality cohort studies in the UK and so if we could coordinate that effort, then potentially we only have to do the applications once and pay once.”

I had not realised that researchers working on studies designed to benefit the NHS had to pay the NHS for access to data which study participants had agreed could be released. But the vision for PC3 is about far more than just saving money - the various datasets follow people from before they are diagnosed right up until they die:

“We have what we're calling the end to end of Parkinson's. And what we want to do with those 10 or 12 datasets is to harmonise the data so that we've got the same metrics and parameters gathered from each of those cohorts - harmonise that all into one data set.”

The next step is to get that giant dataset linked with hospital records: “So things like hospital admission with hip fracture, hospital admission due to infection, deaths, dementia, all these kinds of things. And then we can look at what are the determinants of those and understand more about the trajectories of Parkinson's or the details of things that we want to avoid in Parkinson's.”

Alastair Noyce talks admiringly of a study funded by the Michael J. Fox Foundation called PPMI - “they've got a beautiful dataset for about 1000 patients with Parkinson's’ - but says if all goes well PC3 will have granular data on 10,000 people.

As the project does not involve recruiting new participants you would think getting it off the ground would be simple. Well you’d be wrong - there are a whole lot of bureaucratic hurdles to clear. Take ethics - each individual project had to be cleared by an ethics committee but PC3 will need a new review of its practices before it is allowed to merge the datasets. Then the project may need to get what is known as a CAG - approval from a body called the Confidentiality Advisory Group which looks at access to patient information. And these days any clinical trial or study needs PPIE, a patient and partner involvement and engagement committee, to advise on the direction it should take. The aim is that this will have a particularly prominent role in PE3.

All of this is costly, mainly in terms of employing people to shepherd the project through the bureaucratic assault course which awaits any attempt to do something valuable with patient data. But the prize at the end - a world leading data resource which could help researchers unlock the many secrets of Parkinson’s - is great. So I’m hoping Alastair Noyce and his colleagues get the go-ahead before too many more of my Parkinson’s diagnosis anniversaries have passed.