This guest post was written by Todd Sherer, PhD, the Chief Executive Officer of The Michael J. Fox Foundation for Parkinson’s Research. The Foundation is working hard to direct research that could deliver improved therapies or even a cure to the millions living with Parkinson's disease today. Dr. Sherer also sits on the Advisory Council of the National Center for Advancing Translational Sciences.
If I had a nickel for every time I heard someone (myself included) say “collaboration” since I started at The Michael J. Fox Foundation, I’d live in a much larger house. I fervently believe that partnerships are the path to better treatments and cures for many diseases, including Parkinson’s, and you’d be hard-pressed to find many scientists who disagree. The problem is, we all still work in a system that feeds on secrecy and competition. It’s hard enough work just to dream up win/win collaborative structures; getting them off the ground can feel like pushing a boulder up a hill. Yet there is no doubt that the realities of today’s research environment — everything from the accumulation of big data to the ever-shrinking availability of funds — demand new models for collaboration. Call it “collaboration 2.0.”
Our Foundation has a history of taking calculated risks to try to forge a path that can help define the next generation of scientific collaboration. I share a few recent examples in the hope of increasing the reach of these initiatives, inspiring others like them, and encouraging frank commentary on how they’re working.
Open-Access Data
The successes of collaborations in the traditional sense, coupled with advanced techniques such as genomic sequencing, have yielded masses of data. Consortia of clinical sites around the world are working together to collect and characterize data and biospecimens through standardized methods, leading to ever-larger pools — more like Great Lakes — of data. Study investigators draw their own conclusions, but there is so much more to discover than any individual lab has the bandwidth for.
The open-source model of the Alzheimer’s Disease Neuroimaging Initiative (ADNI) was designed to confront this challenge head-on, and is widely credited with speeding the discovery of Alzheimer’s biomarkers. We paid heed and structured our Parkinson’s Progression Marker’s Initiative (PPMI), launched in 2010 with 13 industry funding partners, to provide similar real-time access for qualified researchers around the world. PPMI has united 24 clinical sites to collect standardized data and samples from an initial group of 423 PD patients and 196 controls, and we’re planning to publish baseline findings this fall on what we think are some exciting leads.
Since its first enrollment, the initiative has offered de-identified clinical data to approved investigators through an online database as soon as it is collected. The aim is simple: maximize the number of minds working on the problem in order to hasten the pace of investigation and drug development. Certainly we’re encouraged by the response of the scientific community: To date, PPMI data has been downloaded close to 75,000 times by researchers around the world, and the initiative has received 25 requests for biosamples. In January a team from UPenn published findings using data culled from PPMI. They found that Parkinson’s itself does not confer a higher risk of impulse control disorders, such as gambling or substance abuse, supporting other research that reports it may be a medication effect that leads to higher prevalence of these disorders among PD patients.
These results made us want to open up real-time access to data from lots of other MJFF studies. So last month, MJFF made clinical data and biospecimens from our LRRK2 Cohort Consortium available to the research community at large. Since 2009, international consortium sites have been studying more than 1,900 Parkinson’s patient and control volunteers to learn more about the most common genetic contributor to Parkinson’s disease discovered to date: a mutation of the leucine-rich repeat kinase 2 (LRRK2) gene. As consortium researchers continue to follow these volunteers, the Foundation will allow real-time access to the information collected in order to glean faster insights into questions of genotype-phenotype associations, clinical features of Parkinson’s when it’s caused by LRRK2 mutations (and how these differ from the more common, sporadic form of the disease), how LRRK2 mutations influence Parkinson’s risk, and other key issues in translating basic genetic discovery into practical, patient-relevant treatments.
We opened access to data and samples from six other ongoing studies at the same time. And since we know the resources don’t help you much if you don’t have funding to do the research, we’re also offering investigative teams awards of up to $250,000 in funding to use our data in high-impact Parkinson’s drug development.
Crowdsourcing
A great way to grow engagement with resources you’re willing to share? Ask for it. Collaboration 2.0 casts a wide net.
We dipped our toe in the crowdsourcing waters earlier this year with our Parkinson’s Data Challenge, which asked anyone interested to download a set of data that had been collected from PD patients and controls using smart phones. We wanted to hear ideas of how teams would use this information in PD monitoring and treatment to illuminate the possibilities with omnipresent mobile devices. After 630 downloads of the dataset from teams in 21 countries, we awarded the $10,000 prize to researchers from LIONsolver, Inc. who provided proof of concept for a machine learning approach that could unveil clues to PD onset and progression embedded in data collected on smartphones. LionSolver is now thinking about retooling their architecture analysis software to the detection of neurodegenerative diseases.
What makes the challenge even more compelling is that the data set cost only a couple hundred dollars to assemble. The researchers bought the smartphones they used from Craigslist for around $50 each and collected the data from participants through email.
A similarly crowdsourced, though arguably less scrappy, effort is under way at The National Center for Advancing Translational Sciences (NCATS), part of the National Institutes of Health, with the $12.7 million pilot initiative “Discovering New Therapeutic Uses for Existing Molecules.” Pharmaceutical companies offered 58 molecular compounds that have already undergone safety testing, and NCATS crowdsourced ideas for use of the compounds. Through this program, nine investigator groups are now working to identify new treatments in eight disease areas.
Cross-Disciplinary Collaboration 2.0
The more we uncover about the interconnectedness and complexity of the human system, the more proof we are gathering that findings and treatments for one disease may provide invaluable insights for others. We’ve seen some really intriguing crosstalk between the Parkinson’s and Alzheimer’s disease research communities recently. At the Alzheimer’s Association International Conference (AAIC) in Boston this past July, researchers discussed new evidence that suggests there may be more overlap in the pathologies of Alzheimer’s and Parkinson’s than once thought. These findings are based on biopsies of patients who were diagnosed with Alzheimer’s, but who actually appear to have dementia caused by the hallmark pathology of Parkinson’s — Lewy bodies created by high alpha-synuclein accumulation in the brain. Continued work across disease areas may allow us to redouble our impact for more patients.
The thirst for data and resources in the scientific community is unquenchable. At MJFF, we have been gratified to see growing momentum around partnerships between key stakeholders across sectors, companies and disease areas. We hope one day that our model of sharing will not be unique. Matt Herper, who lent me the space for this message, says at the top of every page: “I believe this is biology’s century.” And so do I. But believing isn’t enough. We have to take action to conceive, implement and refine new structures for sharing resources and engaging non-traditional partners. The results should be: More ideas. More discovery. Better health.
