ASCO: 'High-Risk' Project 'Must Be Done'

Tech Platform Is Best Known in Finance

Nick Mulcahy

January 22, 2015

The American Society of Clinical Oncology (ASCO) has announced that it will partner with the software giant SAP to develop their "big data" initiative known as CancerLinQ, which the organization claims will "revolutionize" cancer care.

"We believe it can be done and must be done," said ASCO president Peter Paul Yu, MD, about the initiative during a phone conference with reporters yesterday.

However, CancerLinQ is also "high risk," he said, because it is unprecedented among professional medical societies to attempt to build such an extensive "digital health" tool.

Dr Yu, who is from the Palo Alto Medical Foundation in California, said that CancerLinQ will consist of three big informational offerings to clinicians: feedback on how their management of a patient compares with ASCO guidelines; disclosure of patterns in their practice with regard to treatments, patient characteristics, and outcomes; and provision of decision support.

All that help is not cheap.

ASCO has allocated funds in its operating budget in the "eight-figure" range, or between $10 million and $99 million, over a 5-year period. It is anticipated that most of the expenses will be incurred in the first few years, Dr Yu reported.

The organization has also fundraised successfully for the project, said Dr Yu, who declined to specify that amount.

Partnering with a technology company was not part of original plan for CancerLinQ. "We thought...we were going to do this all ourselves," he noted.

However, ASCO officials decided to add a tech partner because "we can't invent everything and there is certainly no sense in reinventing the wheel," Dr Yu explained.

ASCO also announced that eight oncology practices around the United States have signed agreements with CancerLinQ LLC, which is a subsidiary of ASCO, to provide patient records for the "first version" of CancerLinQ. The participants include large well-known organizations such as the Cancer Treatment Centers of America.

Seven more "large cancer centers" will soon be added, and all will be "vanguard practices" in the development effort.

The 15 practices will contribute data on an estimated 500,000 patients, according to ASCO. Individual patients will be asked for their consent and can opt not to participate, said Dr Yu.

Currently, there is no actual working CancerLinQ product.

CancerLinQ will be developed using SAP HANA, a "flexible, multipurpose in-memory data management and application platform," according to ASCO press materials.

SAP HANA is best known as a tool used by the financial industry. Whipping it into healthcare shape will be a big job, said Dr Yu.

"We're talking about a major development project over several years to take this market-leading product and bring it to a whole new industry," he explained.

ASCO pays SAP a licensing fee to use SAP HANA, but the technology company will also invest its own money in developing the needed software and databases.

SAP HANA has been used in other healthcare projects, ASCO press materials report. The National Center for Tumor Diseases in Heidelberg, Germany, uses the platform "to help accelerate cancer research and improve clinical trial matching," and the Mitsui Knowledge Industry in Japan uses it to "reduce patient-specific genomic analysis for cancer diagnosis from 30 days to 20 minutes."

CancerLinQ will be free to ASCO members and their practices and institutions, said Dr Yu.

ASCO has competition in the race to develop a functional big data tool. Private companies, such as Flatiron Health in New York City, which is partly funded by Google, and McKesson, are also developing tools.

Not everyone in the healthcare field believes big data efforts such as CancerLinQ are revolutionary.

Robert Weinberg, PhD, from the Massachusetts Institute of Technology in Cambridge, said in an interview earlier this year that there has already been a great deal of mining of clinical cancer data, but "relative to the effort that's been put into it, there's been little in take-home lessons" for clinicians.

Dr Weinberg is considered to be a major figure in cancer research and is credited with discovering the first human oncogene, RAS, and the first tumor suppressor gene, Rb.

"The promise of big data has been greatly oversold," Robert Carlson, MD, chief executive officer at the National Comprehensive Cancer Network, told Medscape Medical News in 2014.

Dr Carlson explained that one of the justifications for large-scale data aggregation is that it is needed to make sense of the burgeoning information on the molecular and genetic characteristics of cancers.

But the clinical importance of the molecular characteristics of tumors has only been established in a few specific cancer types, he pointed out. Most of this kind of data is currently inconsequential in the clinic, he said.

Big data has also been criticized as being observational data, for the most part. In other words, big data comes from patient record databases and is not the product of a randomized controlled trial, the gold standard of medical evidence.

Critics observe that there is already a lot of observational data that has been collected in oncology, especially in Surveillance, Epidemiology, and End Results (SEER)–Medicare databases, which is regularly mined by researchers for clinical insight.

The SEER–Medicare project is sponsored by the National Cancer Institute (NCI), and has details on 1.6 million cancer patients and their treatments, outcomes, and second cancers, among other data, according to an NCI report.

However, Therese Mulvey, MD, a medical oncologist at the Southcoast Centers for Cancer Care in New Bedford, Massachusetts, which is collaborating with CancerLinQ, said the ASCO effort is a big step toward improving observational data.

Other observational data mining represents "little bites at the [medical] chart apple," she told Medscape Medical News in an interview.

Dr Mulvey, who is a member of the ASCO board of directors, explained that sources such as SEER–Medicare have limited data. Medicare does not have recurrence data, for example.

Furthermore, information on important clinical matters, such as patient pain, cognitive function, and activities of daily living, are missing from major observational datasets. In short, the ambitions of the ASCO data project include patient data that are currently not collected and analyzed in a big way.

Dr Mulvey is a believer in the importance of big data and ASCO. "I'm convinced ASCO can do this," she said. "ASCO has the trust of doctors."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.