NeuroTalk Support Groups - MJFF Recruit 10,000 for Web-based PD Research

Sorry...I was on the road for a few days so I'm just getting to this...

I think there are layers and layers to what we can learn (so, basically how data can be used) when studying people who have the disease (as well as those at risk for developing the disease). We do some of this now, with more traditional approaches and our hope is that we can do the same, if not more, with this new approach over time.

So, where to start...

How will data be used? Aggregation of data in large numbers might take us to new, more relevant findings.

The goal of clinical studies generally is to collect data and make observations and inferences. Validity of observations and inferences are often hampered by the fact that we have to study vast numbers when we know / expect that the observations will be rare (so, observing enough patterns to tell us about the causes of PD when it is generally seen as sporadic means many, many people need to be studied). Separately if we want to observe something that is gradual, it means we need to study over a long period of time. Both of these needs lead to great expense. Historically some of the most interesting observations about human disease and risk factors have come from long term (decades long) epidemiology studies. And, these studies are labor intensive (either lots of clinic hours or even folks how go out into the community to collect data. Again, read expensive.

So, in general, imagine if you could harness the power of the internet to decrease costs and speed up intervals of data collection….maybe we could collapse the time frames and maybe we could expand the number of people we could study…pretty tempting since right now we basically draw inferences from relatively small numbers of people over shorter than desired periods of time. Maybe we will finally get significant data that could point to subtypes of disease or help us predict progression patterns for different patients…all stuff that we currently don’t know and might not be able to ascertain with current methods.

What is intriguing about this community is that we can take these hypotheticals out for a test drive.

In this 23andme community, the goal is to collect data on a large numbers of patients with Parkinson’s disease (10,000 to start). The initial data will be genetic (from the spit test) and over time behavior and environmental data will be overlayed (we have been funding a collaboration between the Parkinson’s Institute and 23andme to see if they can validated online surveys that could be reasonable substitutions for the kind of data traditionally collected in doctor’s offices or in phone surveys.

Some other elements of this community are intriguing departures from traditional approaches. In the US (although not so in many other countries) when a patient signs up to participate in a clinical study (to look say for genetic markers of PD) the patient is not made aware of the results. This is a controversial point. Perhaps thousands of PD patients have donated blood to studies before and while the researcher knows the results of the particular patients’ genetic status, the patient does not. In this 23andme effort, not only to all participants know about the PD-related genes they report on but the participants will learn about genetic risk factors for a list of over 100 diseases (and that number will grow). This particular piece is a hallmark of the 23andme philosophy—patients have a right to know and control their data...it is cheer for patient empowerment. But it is not without controversy. Doctors and IRB’s generally feel that it is unwise to tell people about their genetic risks factors for disease if there isn’t an approved treatment to offer with the news. By the way, some people also make a personal decision that is similar…I don’t want to know information if I can’t do anything to change it. This is one of the reasons you see the disclaimers about how important it is to consider the consequences of getting access to your personal genetics information and why we stress it in our communications about this community. But, ultimately, we fall on the side of making the offer to people to decide for themselves (which isn’t the case when you participate in a more traditional study where such information might be learned, but not shared). So, I guess this is another why in which data will be used…it can/will be used by patients themselves.

The community is an interesting twist too…with your own data in hand, you can choose to share (and compare) with others in the study (this is at your discretion). You can also present ideas to the 23andme staff—it is an interactive community. They intend to continue surveying the community over time and, importantly to allow other researchers to survey the community over time…and, which each set of answers, the database has the correlated genetic analysis … so, basically pairing genetics and phenotypes in one place as a foundation for research. This would be a significant departure from the currently siloed data stored and coveted in spreadsheets at various research institutions around the world. It is important to note that 23andme won’t be sending around individuals’ genetic data to various researchers but rather providing access to their community…so, let’s say there is a new hypothesis that lifelong gum chewing is neuroprotective, a scientist could ask 23andme to poll their community about gum chewing…

Who’s doing the conceptual work? So, some work is being done by 23andme (expanding the number of genes in PD that are reported on –to their PD community and beyond or can they collect/store data about symptoms changing over time and correlate it to diary entries about meds—I’m making this up but basically they can add features/dimensions to the nature of data collected). Some work is also being done in partnership with clinical study experts (such as in the MJFF-funded project with Carlie Tanner at the PI)…some will be done overtime as various experts develop new ideas/queries (gum chewers?)…

The overriding elements today are, however, the possibility of getting to large numbers quickly and cost-effectively (of course costs here are covered by Sergey Brin’s grant to 23andme) and the fact that patients would have unprecedented access to their own data and ongoing mechanisms to contribute additional data to the study population.

We don't have all the answers yet as this is admittedly a new strategy/approach...but, certainly what intrigues us is the possibilities that could come from novel ways to collect data.

Debi