by Kate Ackerman
iHealthBeat Managing Editor
Experts say an announcement made last week by the American Society of Clinical Oncology could have significant implications for the future of health care delivery.
ASCO announced that it successfully completed a prototype of a cancer care database that leverages health IT to improve care.
CancerLinQ is a knowledge-generating computer network that will collect and analyze cancer care data from millions of patients and then combine that information with expert guidelines and other evidence to generate real-time, personalized guidance and quality feedback for doctors.
ASCO officials demonstrated the prototype at a briefing at the National Press Club last week. It includes de-identified data from more than 100,000 U.S. breast cancer patients.
Clifford Hudis, an attending physician and chief of the Breast Cancer Medicine Service at Memorial Sloan-Kettering Cancer Center and ASCO president-elect, referred to the prototype as “a proof-of-concept,” saying that it shows the group’s vision is possible.
Details of CancerLinQ
To develop the CancerLinQ prototype, ASCO relied on open-source software whenever possible, Hudis explained.
The system is able to collect patient data, such as laboratory tests, genomic profiles and physician notes, in any standard directly from any electronic health record system.
CancerLinQ features four core functions:
- Real-time data collection;
- Clinical decision support;
- Data mining and visualization; and
- Quality feedback.
How CancerLinQ Could Change Medicine
Sandra Swain, medical director of the Washington Cancer Institute at Medstar Washington Hospital Center, a professor at Georgetown University and ASCO president, noted that only 3% of cancer patients participate in clinical trials and that older patients in particular are left out of such research.
Swain said that CancerLinQ will “help us unlock that 97% of data that’s beyond our reach,” adding that the “result will be higher quality, higher value care and better outcomes for patients.”
Charles Penley, a partner at Tennessee Oncology and chair of the Conquer Cancer Foundation Board of Directors, explained that when he encounters a puzzling case, he turns to the literature or contacts a few colleagues to ask them if they have any insight.
But CancerLinQ could change that. The new tool gives health care providers the “opportunity to query not just a few experts you know but the collective experience of hundreds of thousands to millions of patients,” Penley said. “This will be second opinion times multiples.”
John Sharp, manager of Research Informatics at the Cleveland Clinic, called CancerLinQ “an ambitious attempt to achieve a piece of the learning health care system,” adding, “Rather than developing treatment guidelines over years of clinical trials, the learning health care system proposes a more rapid cycle of analyzing real-time data and incorporating this into evidence-based guidelines.”
Lynn Etheredge, director of the Rapid Learning Project at George Washington University, first proposed the concept of a rapid learning health system in a 2007 Health Affairs article.
He wrote, “Private- and public-sector initiatives, using electronic health record databases from millions of people, could rapidly advance the U.S. evidence base for clinical care. Rapid learning could fill major knowledge gaps about health care costs, the benefits and risks of drugs and procedures, geographic variations, environmental health influences, the health of special populations and personalized medicine.”
Etheredge, who sat on an expert panel at ASCO’s briefing, said he was pleased to see some of his predictions come to life.
“It’s like talking about going to the moon. Someone’s got to do it. Someone’s got to lead,” he said, adding that ASCO deserves “enormous credit” for taking the initiative.
Etheredge said, “This is physician leadership for improving health care,” something that’s been missing for the past 20 years.
Hudis said that CancerLinQ will allow health care providers to tap into the collective experience of millions of patients to inform care delivery or to provide insight into where resources should be spent on research.
Next Steps
Hudis said that while the development of a successful prototype is “an important milestone,” it’s the first step in a long process.
ASCO plans to further evaluate the prototype and release a series of white papers on the lessons learned, he said.
Expanding the project to all types of cancers will require significant time, resources and funding, Hudis said.
Sharp noted that there are several barriers to expanding the prototype to a more general data extraction for other cancers.
He said, “The first challenge is getting general agreement to share de-identified data from dozens of cancer centers. One of the biggest challenges in data sharing is not the technology but the legal agreements even with de-identified data.”
Sharp added, “Second, de-identification of [EHR] data is a challenge as many medical informaticists know. Complete de-identification includes date-shifting or removing dates, which adds unique challenges for data analytics.”
Cost and sustainability could be another barrier, he said. “Many data sharing initiatives have been supported by grants and lack sustainability without government or industry support. Fortunately, ASCO is looking at this problem as part of the initiative.”
Swain stressed the need for collaboration, saying that ASCO wants to work with the larger health care community on the initiative.
“This is a hugely ambitious project, it’s a very bold project, it’s a scary project,” Swain said, adding that she believes it ultimately will be successful because of physician leadership.
“If this is successful as a lightweight data-sharing model, all other areas of medicine could benefit,” Sharp said.
He said, “Existing national registries, such as those of the Society of Thoracic Surgery, have required significant resources by member institutions to validate and maintain the data,” adding, “A new approach incorporating rapid data acquisition, analytics and guideline recommendations would be transformative.”
Sharp said, “There will be many critics of this approach both in terms of trusting the data, which [are] not acquired in highly structured clinical trials, and general distrust of the accuracy of [EHR] data. But as this initiative is envisioned as breaking through barriers, I think it is a candidate for being a game changer.”
This article was originally published on iHealthBeat.org and is used here with permission.