Every year, over half a million Americans die of cancer. That’s roughly 1,500 people per day. Each instance of cancer can be unique, and a treatment that may save one patient may not work for another. Only recently have researchers and doctors become able to use genetic and other data to design patient- and tumor-specific treatments – a breakthrough that can depend on the ability to manage and interpret massive amounts of data.
How today’s electronic medical records fall short
The American Society of Clinical Oncology (ASCO), a professional organization of oncologists, is on the brink of translating huge amounts of raw cancer data from various types of electronic medical records (EMRs) into knowledge that can be applied to help future patients. In my time at ASCO, I’ve seen the transition from paper-based medical records to EMRs. On the surface, that in itself is a great achievement. However, the first generation of EMR programs and systems were not designed to be integrated and interoperable, but instead to address service and billing precision. Researchers now have a huge repository of cancer care data – but this data is nonstandardized and difficult to translate between systems. Hence, while the opportunities this data presents are great, many of the inefficiencies of the paper-based system remain.
Data locked up in too many systems
Medicine is poised to embrace powerful data insights and improve treatment and outcomes, but our care delivery technology hasn’t kept pace with this dream. Big retailers can mine their customers’ activity to recommend the best winter coat or a popular travel destination likely to appeal to a shopper, but more important medical goals remain unmet. We need to catch up.
The volume of existing data on both cancer research and specific patients is immense – but unstructured. Once we can access, search, and utilize that wealth of information, how much progress could we make toward stopping this disease? Today, only 3% of cancer patients participate in clinical trials, with the resulting findings used to develop treatments for the general patient population. What advances in care could be made if we could centralize, access, and analyze data for even half of the remaining 97% of patients?
Tapping the potential of data to save lives
ASCO is leading the development of a system, CancerLinQ, that can extract, aggregate, and mine cancer data from a wide range of sources. The initiative supports ASCO’s mission to deliver quality care to a broad range of patients by rapidly extracting and learning from everyday records. The timing is also excellent, in that the project supports Vice President Joe Biden’s call for a “cancer moon shot” and the Obama-Biden administration’s Precision Medicine Initiative, whose goal is to “enable a new era of medicine through research, technology, and policies that empower patients, researchers, and providers to work together toward development of individualized treatments.”
When available, CancerLinQ can be used to:
- Help identify drug outcome associations
- Replace cumbersome manual data extraction by automatically mining data in real time
- Provide doctors with meaningful clinical and decision support
The evolution of CancerLinQ
ASCO began building CancerLinQ with off-the-shelf open-source software, but quickly realized that the project required more powerful technology and software expertise. We decided to bring in SAP, leveraging its technology, analysis skills, and Big Data expertise. The resulting collaboration with our wholly owned nonprofit, CancerLinQ LLC, has dramatically accelerated the development of CancerLinQ. Today, the solution has been successfully tested on a population of a million (manufactured) patient records. The SAP and CancerLinQ teams are now populating it with live patient records, with the goal of incorporating data from a million records by summer 2016.
Rapid insights that benefit patients
CancerLinQ‘s innovation helps unlock knowledge from the medical records of the 97% of cancer patients not involved in clinical trials, enabling better, more data-driven decision making. Doctors deciding on treatment plans can get new insights in seconds, not years, and tailor cancer treatments to patient needs throughout the treatment process. Ultimately, the goal is to support better patient care and quality of life, incorporating both insights drawn from Big Data and patient-reported outcomes (including from healthcare wearables).
IT promises to be an agent of change, accelerating learning, reducing healthcare costs, and engaging the health ecosystem of patients and families, researchers, payers, and regulators. CancerLinQ, a rapid-learning health system for oncology, can help lead this effort.
CancerLinQ runs on the SAP HANA platform and uses Big Data analytics to unlock real-world patient data from EMRs. Each iteration of CancerLinQ will deliver more-powerful tools and insights to physicians, researchers, patients, and others in the cancer community. Learn more about CancerLinQ’s sophisticated platform, and how SAP Foundation for Health can help unlock biometric big data with advanced analytic solutions, at SAP Personalized Medicine or continue the discussion on Twitter @SAP_Healthcare.