Big data essential to cancer 'moonshot'
In remarks here at the annual Health Datapalooza conference, Biden said that it is essential for the various organizations focused on cancer research and patient advocacy to work more closely together, pooling data and resources to advance what the administration has termed the "cancer moonshot."
Likewise, he stressed the potential for massive sets of genomic data -- and the computing power to analyze them -- to develop novel treatment programs that would boost remission rates or extend life expectancies and improve the quality of life for cancer patients.
[ Related: How CIOs can prepare for healthcare data tsunami ]
"There's a growing recognition of the need for increased collaboration among the private sector, academia, patient foundations and government," Biden said. "And there's an acknowledgement that big data and the computing capability we possess today -- if we cooperate -- could help us understand each person's cancer and uncover similarities and responses across large patient groups and help us design the best course of therapy."
Biden said that leaders of countries around the world have expressed considerable interest in the administration's cancer initiative, with some looking to develop memoranda of understanding to join in the effort.
The technologies that will be critical to accelerating the fight against cancer are similar to the data-driven applications and analytics that tech firms have been pitching to businesses in industries ranging from retail to energy.
[ Related: Experts urge caution when putting health data in the cloud ]
"Big data captures the big picture, the complexity, the challenges and the opportunities in treating cancer. It matters a lot, because cancer is a disease of the genome, and genomic information can help give us therapeutic roadmaps," Biden said. "Big data and computing power together provide the possibility of significant insight into how genomics, family medical history, lifestyles, genetic changes can trigger cancer and how the cancers can be treated. Why does one therapy work on one out of a hundred or three out of 500 patients and not the rest"
But Biden acknowledged that significant roadblocks stand in the way of his vision of tapping into massive data sets to advance the medical community's understanding of how cancer develops and spreads, and how best to respond to it.
"[I]n order for this promise to be realized we have to do a few things," he said. "First, we have to generate enough data to qualify as big data -- millions of these samples, and it has to be data that's readable and usable. That means scientists all over the world using powerful computers need to be able to access, search and analyze the data."
So in order for researchers around the world to make use of the cancer datasets that Biden describes, the information must be in a standardized format that high-performance computers can understand.
[ Related: How to balance data privacy and healthcare improvements ]
Likewise, Biden stressed the importance of reforming a culture pervasive in the healthcare community that tends to resist sharing data in favor of keeping information assets under lock and key.
"The second thing we have to do is we have to share the data. We have to share it in order to aggregate large amounts of data that can be used to find solutions," he said. "Today, different technology systems can't talk to each other. Most major cancer centers don't have an easy way -- and in many cases the motivation -- to share data including patients' records, test results, family histories, and treatment responses. We've got to change this. This information, this is information that will allow physicians, including community oncologists wherever they are located, to make care decisions for their patients based on the best information and treatments as rapidly as possible."
In the same vein, Biden called for the research community to abandon the practice of detailing new and potentially breakthrough findings in journals that live behind expensive paywalls.
"Researchers need to share data in order to move discoveries more rapidly," he said, calling on the scientific community to join in a collaborative effort to cut down on "unnecessary duplication and failed efforts and wasted time and money" and affirming the limits of what the public sector can do by way of incentives and regulations.
"While our government can do a great deal," Biden said, "this is not the work of government alone."