Industry Voices: Researching ALS through new eyes–Why data science will foster life science breakthroughs

Amyotrophic lateral sclerosis, commonly known as ALS, is an enduring mystery for life scientists–even those who have dedicated their careers to understanding it and finding a cure. Terminal, essentially untreatable, and frustratingly hard to predict, the disease has eluded us for years–and has achieved a sort of “orphan” status that makes it hard to draw attention and research dollars to ALS.

But an innovative new approach, drawing on the disciplines of both medical science and data science, offers the promise of an unprecedented breakthrough in the search for a cure. For the first time, Big Pharma and Big Data are coming together to foster insights in ALS. Specifically, working with nonprofits like mine, Teva ($TEVA), Novartis ($NVS), Sanofi ($SNY), Regeneron ($REGN),Knopp Biosciences, and other companies have opened up their clinical trials databases to help create the Pooled Resource Open-Access ALS Clinical Trials (PRO-ACT) database, an open-access repository of ALS patient and clinical trial data made available to researchers, patients, and members of the public. Although recent efforts by Big Pharma’s two major trade groups, EFPIA andPhRMA, will increase the amount of clinical trial data available in the future, PRO-ACT has already vastly increased the breadth and depth of ALS clinical data available.

Why ALS needs an open-access database

But just because we’ve built it, does that mean the researchers–and the breakthroughs–will come?

We think the answer is “yes.” Here’s why: Previously, access to ALS clinical trials data was limited and piecemeal. Scientists interested in seeking a better understanding of past drug trials were frequently limited to a single trial or a handful of small studies, and even these data were often limited to placebo arms or a small subset of measures. This lack of data access meant that important questions about disease natural history and disease heterogeneity were unanswerable. Clinician-scientists interested in designing clinical trials of novel treatments were operating with the equivalent of one hand tied behind their backs with all of that valuable data out of reach.

Recognizing this missing link, we reached out to major biotech companies around the world and asked them to donate their ALS data in order to develop PRO-ACT, an important open-access database of thousands of clinical trial patient records. The database is anonymous–identifying patient information isn’t a part of the picture. Rather, the database serves as a unique resource to approach ALS research in a completely new way. By reimagining ALS as a problem to be tackled not only by life scientists but also by data scientists, we have been able to foster the creation of a prediction tool that shows real and measurable promise for the disease.

Crowdsourcing breakthroughs

Just as crowdfunding can fund a new startup or an emerging artist or more efficiently lead to the development of new software products, making a large-scale clinical trials database available to a large number of analytical thinkers from a variety of disciplines opens up new possibilities for insights and breakthroughs.

To get PRO-ACT off the ground, we crowdsourced ideas from scientists and mathematicians around the world, and offered a financial incentive to whomever could develop an algorithm that would help predict disease progression–a major stumbling block for clinical researchers, who have struggled to design effective trials in part due to ALS’ unpredictable nature. Some patients live mere months, while others live for years or even decades after their diagnosis.

By crowdsourcing a “Big Data” solution, we were able to engage an entirely new set of researchers, data experts, and good old-fashioned math whizzes in solving a problem that has stymied researchers for years. Using the power of machine-based learning (i.e., the ability to repeat a task millions of times without fatigue), these quantitative experts were able to sift through millions of data points in order to identify commonly collected disease measures with the power to, in combination, distinguish rapid from slowly progressing patients early on in the course of their disease. In the process, they were also able to exclude a host of things that surprisingly didn’t seem to robustly predict disease.

Any scientist will tell you that sometimes it takes a change in perspective to be able to see–and start to solve–a problem in a new way.

Big Pharma, meet my friend, Big Data

The pharmaceutical industry has certainly stepped up its efforts recently to court and cultivate data science, and we’ve seen encouraging signs that companies are increasingly using innovative approaches to glean insights from information like that contained in PRO-ACT. For example, almost a dozen biopharmaceutical companies have downloaded PRO-ACT data so far–a strong indicator of these companies’ interest in mining new data to “crack the code” of ALS.

In their efforts to understand Alzheimer’s disease, the Critical Path Institute used a similar Pharma-donated “Big Data” collection effort to build a clinical trials simulation tool that was recently approved by the FDA as a “fit-for-purpose” drug development tool for Alzheimer’s disease.

From open mind to open access

As a scientist, I have been trained to approach problems with a methodology and process that yields the greatest chance of a result that will answer my desired question. But what about the question(s) I haven’t thought to ask?

Scientific discovery is nearly always made better by collaboration and sharing–as with everything else in life, a second opinion, alternative perspective, or new way of thinking about a problem can lead to better results.

Technology industry trends such as crowdsourcing and Big Data analytics have fostered a lot of new thinking in the scientific world, and I think it’s safe to argue that this new “open-access” approach is better for the science and, ultimately, better for patients.

When I open up my data to everyone, the biggest risk I run is that someone else might identify a pattern or discover a relationship between data points that I haven’t thought of. What a great problem to have, and a risk well worth taking.

Read more: Industry Voices: Researching ALS through new eyes–Why data science will foster life science breakthroughs – FierceBiotech http://www.fiercebiotech.com/story/industry-voices-researching-als-through-new-eyes-why-data-science-will-fost/2013-08-26#ixzz2dcrfGocG
Subscribe at FierceBiotech

Advertisements
This entry was posted in Uncategorized and tagged , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s