Most knowledge about what works and what doesn’t in cancer treatment draws from the meager 3 percent of cancer patients who participate in clinical trials. A new partnership between the American Society of Clinical Oncology (ASCO) and SAP, the German software giant, looks to leverage “big data” to glean insights from the other 97 percent.

The ambitious project, called CancerLinQ, has been in development since 2012. Late this year eight community oncology practices and seven large cancer centers will roll it out, placing the electronic health records of around half a million cancer patients at the disposal of oncology practices across the U.S.

The goal is to eventually tap into the records of nearly every cancer patient in the country, project leaders announced at a Jan. 21 press briefing.

Peter Yu, MD, director of cancer research at the Palo Alto Medical Foundation in California and president of ASCO, said CancerLinQ will allow physicians to compare their actual care against guidelines, feeding cancer docs “individualized and unbiased” decision support for every patient and every type of cancer.

“When CancerLinQ is complete, doctors will gain insights in seconds, not years,” he said. “And patients will gain by having access to high-quality care based on the most up-to-date findings.”

Yu declined to specify who is paying for the project, but allowed that ASCO has allocated dollars “in the eight-figure range” in its operating budget over a five-year period.

CancerLinQ will maintain control over all data, services, products and everything that stems from the initiative, including clinical decision support tools and analyses, said Clifford Hudis, MD, chief of the breast medicine service at Memorial Sloan Kettering Cancer Center in New York and ASCO’s immediate past president.

Hudis said SAP will provide access to its widely used HANA database-management system, including tools that are customized to CancerLinQ’s unique needs. “We aim to take an unprecedented amount of patient data, identify meaningful and actionable insights and improve patient care,” Hudis said. “And we aim to do this faster than anybody has before or is planning to.”

Hudis added that CancerLinQ is one of the only major cancer data initiatives developed and led by physicians with the primary purpose of improving patient care.

Therese Mulvey, MD, physician-in-chief at the Southcoast Centers for Cancer Care in New Bedford, Mass., one of the original eight to adopt CancerLinQ, said she hopes to use the CancerLinQ platform to move quality improvement efforts “from process to outcomes.”

“Many patients have molecular markers performed on their tumors, and the state is locked away and unavailable for comparison,” added Mulvey, pointing out that there are more than 13 million cancer survivors in the U.S. “What happens to them after they complete their therapy will be an area of great interest to us practicing in the community and caring for those patients.”

imagingBiz asked the panel, which also included David Delaney, MD, chief medical officer of SAP America, what particular benefits or opportunities CancerLinQ presents to diagnostic radiologists.

“As we look into outcomes measurements—does this treatment work and who does it work for?—a key component of that is imaging,” said Yu. “That’s a large part of the evidence that shows you’ve actually impacted outcomes.” He said ASCO signed an agreement to work together with the College of American Pathologists about a month ago and is “looking to talk to the radiology societies and the radiation oncology societies as well.”

Will CancerLinQ’s decision-support tool incorporate an imaging-specific component?

“In our [implementation of the] Choosing Wisely program, we looked at some of the imaging tests that really don’t add value to the cancer patient and should not be done routinely,” replied Yu. “But it’s also very important to say when things should be done. When new technology comes along, which is one of the hallmarks of oncology—we’re always getting new technology—what is the best way to use that new technology? How can we find that out more rapidly not just from clinical trials but actually looking at the real world? The element of rapid-learning healthcare systems is a large part of the inspiration behind CancerLinQ.”

The project was not met with pure enthusiasm from all quarters of the oncology community. Big data is “not going to be the holy grail,” Robert Carlson, CEO of the National Comprehensive Cancer Network, told the Philadelphia Inquirer. “There’s a challenge not only in collecting the data—you also need statistical analytical methods that are standard and validated. It’s not clear to me that we have those.”

Meanwhile, however, the CancerLinQ rollout ended up dovetailing with President Barack Obama’s announcement of a new “precision medicine initiative” during his State of the Union address on Jan. 20. The goal, said Obama, is to “bring us closer to curing diseases like cancer and diabetes — and to give all of us access to the personalized information we need to keep ourselves and our families healthier.”

“Big data that is locked in electronic medical records or in molecular diagnostic labs have tremendous potential in health care,” Yu said less than a day later. “We’re all excited about that vision.” 

ASCO has put up a CancerLinQ website.