It was an ambitious undertaking for any organization, let alone a nonprofit: to develop a real-world database of patients with cancer, with data pulled from patient electronic health records and then deidentified and aggregated for use among physicians and researchers to glean insights into ways to improve cancer care in the practice and across the oncology community. 

A decade later, use of real-world cancer care data has become established, and ASCO’s mission-driven commitment to CancerLinQ is paying off for its members, for patients, and for the greater cancer community. CancerLinQ today is helping improve the quality of cancer care by capturing and learning from the real-world experience from diverse patients with cancer across the U.S. Further, many of the technical solutions, guardrails, and frameworks we developed and pioneered within and adjacent to CancerLinQ have become incorporated into broadly accepted standards. 

With these changes in the oncology data landscape and the need for more resources to continue to build out CancerLinQ, the time was right to transition its ownership to a different kind of organization that can take the mission of CancerLinQ to the next level and to an even broader audience. 

On December 11, 2023, ASCO announced that it was transferring ownership of CancerLinQ LLC to ConcertAI, a long-time CancerLinQ vendor and customer with capabilities in real-world data, analytics, next-generation AI, and clinical trial facilitation. ConcertAI has been involved with CancerLinQ since late 2017, when Precision Health AI—a predecessor of ConcertAI—signed an agreement with ASCO. 

In the estimation of ASCO CEO Clifford A. Hudis, MD, FACP, FASCO, “This transition is a powerful acknowledgment of CancerLinQ’s value and its even greater potential to improve patient care. The change in ownership will provide resources to build on CancerLinQ’s successes to date and support its long-term sustainability, both of which require capital on a scale that is available primarily through an investor-backed enterprise like ConcertAI. We expect that ConcertAI’s existing knowledge of and commitment to CancerLinQ will provide for a smooth transition and enhance value to participating practices.” 

In a complementary agreement that was written in conjunction with the sales agreement, ConcertAI has committed to expanding CancerLinQ’s data and network and making sure datasets are available for researchers that ASCO may designate or that are ASCO-affiliated.

“There’s no loss at all—if anything there are gains in the quality of the characteristics and scale of the data that ASCO has available for its own research. This is more on a continuum than a discontinuity for CancerLinQ,” ConcertAI CEO Jeff Elton, PhD, stated. 

Where It All Began 

In Dr. Hudis’ mind, the CancerLinQ journey began with the American Recovery and Reinvestment Act of 2009, which included provisions for transitioning medical records from paper to digital with interoperability as a focus area—potentially enabling different electronic health record (EHR) systems to communicate with one another—for information to be compliantly transmitted and shared “with underlying hope that there would be accelerated learnings from all of that,” Dr. Hudis noted. 

Around that time there was also an Institute of Medicine (now the National Academy of Medicine) report that described and called for the development of what were called “rapid learning systems” across medical disciplines, with many papers being published that outlined a framework and visualized what these systems would be like. 

The ASCO Board of Directors and leadership took this as a challenge to stand up what was originally thought would be a family of rapid learning systems, which for ASCO would be centered on oncology. 

“The very simple vision at the beginning of this was that we would tap into these newly created and rapidly expanding resources, the electronic records systems, and we would bring all that data in, consolidate it, and then explore it to accelerate learning and to help our members and our broad community recognize opportunities to more rapidly increase the quality of care they delivered,” explained Dr. Hudis. 

Building the Prototype 

The early challenges were centered on getting data onboard from medical record systems. 

One of Dr. Hudis' favorite anecdotes that exemplifies the challenges of working with EHRs is that in 2013, in the prototype CancerLinQ system with its first 1,000 or so records from eight medical record systems, staff identified more than 60 ways that the simple reporting of white blood cell count was recorded. For the data to become unified and usable, staff had to transform each of the different versions of that one line item, white blood cell count, into a single representation to put it into a standard database for exploration. 

The team persisted, and a CancerLinQ prototype was first unveiled in spring 2013 at an ASCO-hosted panel discussion on big data in cancer care held at the National Press Club and then at the 2013 ASCO Annual Meeting. The prototype demonstrated, for the first time, the feasibility of a health information technology (HIT)-based learning health system as described above. This first iteration of the CancerLinQ database would include data on more than 133,000 cases from oncology practices across the country. The goal was to publish the lessons learned over the next year and use them to inform the development of the full CancerLinQ system. 

In 2015, the first version of CancerLinQ was rolled out with patient records from 15 vanguard oncology practices, representing the data of approximately 500,000 patients. The plan was for the insights gathered from this data to help lay the groundwork for developing the full system in which oncology practices across the U.S. could participate. The first eight of these institutes included large cancer centers, such as Cancer Treatment Centers of America and Inova Comprehensive Cancer & Research Institute (now the Inova Schar Cancer Institute), and smaller community practices, such as Marin Cancer Care and Michiana Hematology-Oncology, P.C. 

First Collaboration With ConcertAI’s Predecessor 

At the end of 2017, ASCO entered into an agreement with the data services companies Tempus and Precision HealthAI, a predecessor entity now part of ConcertAI. The goal of this arrangement was to accelerate the development and delivery of high-quality, de-identified clinical databases for use by CancerLinQ, its oncologist subscribers, partners, and the broader cancer community. It also enabled CancerLinQ to build and expand on its core data sharing and quality improvement platform for oncology practices more quickly while bolstering Tempus and ConcertAI connection to industry to support faster and better development of new treatments for patients. 

Through the agreement, Tempus and ConcertAI curated the CancerLinQ data into high-quality datasets. These datasets were made available through CancerLinQ Discovery, a service that provides access to high-quality, de-identified datasets for nonprofit organizations, health care providers, academic researchers, and government agencies. 

Tempus and ConcertAI then brought these de-identified datasets to industry partners to produce practical applications such as targeted therapies, more efficient clinical trials, and data-driven analyses to help determine which therapies are producing the greatest patient benefit. CancerLinQ’s proceeds from the license were reinvested to further develop CancerLinQ solutions. 

Setting the Path to QOPI Certification and ASCO Certified With SmartLinQ 

In June 2020, CancerLinQ LLC launched the SmartLinQ QOPI Certification Pathway, an application that enabled oncology practices to automate quality measure tracking and reporting for participation in ASCO’s Quality Oncology Practice Initiative (QOPI) Certification Program, a 3-year certification recognizing practices’ commitment to high-quality care for outpatient oncology practices. 

The SmartLinQ QOPI Certification Pathway reduces staff time and expenses for quality reporting for QOPI Certification by securing taking in, harmonizing, and codifying data from practices’ EHR systems. One of the SmartLinQ QOPI Certification Pathway pilot practices, New Mexico Cancer Center, saved approximately 90 hours of staff time when using SmartLinQ to automate its recertification for the QOPI Certification Program. 

SmartLinQ also has allowed CancerLinQ practices to automate their quality activities, derive actionable insights from real-world patient data, and benchmark their care against the CancerLinQ network. Practices are able to track their progress daily against a broad set of quality measures, both for the practice overall and for specific sites and individual physicians; to benchmark their care against that provided by their peers; to identify “actionable patients” who still needed specific care or treatment to comply with quality standards; and to identify and fix data quality issues, such as missing or inaccurate data in patients’ EHRs. 

ASCO will be migrating its QOPI Certification Program to a new initiative, ASCO Certifiied, which certifies oncology group practices and health systems that meet a single set of comprehensive, evidence-based oncology medical home (OMH) standards from ASCO and the Community Oncology Alliance (COA). These standards focus on seven domains of cancer care: patient engagement; availability and access to care; evidence-based medicine; equitable and comprehensive team-based care; quality improvement; goals of care, palliative, and end-of-life care discussions; and chemotherapy safety. By participating in the program, practices can identify and provide the best possible cancer care to their patients while also driving greater value for the health care system. 

ConcertAI will be working to add more of CancerLinQ’s participating practices onto the QOPI Certification Pathway and then aiding the evolution to ASCO Certified using the SmartLinQ tool for the required quality measure reporting. 

The Future Will Be Prospective Data 

While all the platform enhancements, increased capabilities, and collaborations were taking place, CancerLinQ also was increasing its database of de-identified patient data, bolstering the rich pool of data from which insights are gleaned. 

The papers and abstracts that have been generated to date have been using retrospective data that CancerLinQ has generated—EHR data from past clinical encounters have been aggregated and the de-identified data used to draw conclusions. 

Dr. Elton believes that the future is in prospective studies, which is where individuals are followed over time, and data is collected as their circumstances change. 

Prior to the transfer of ownership, CancerLinQ was in the early stages of developing a product to facilitate clinical research dubbed TriaLinQ. The first feature was to use artificial intelligence (AI) to help identify patients eligible for trial enrollment. ConcertAI has a similar system already in use at a number of cancer centers and oncology practices in the U.S., so the company will use these foundations to build out more research tools, including the ability to automatically populate electronic case report forms directly from EHR data, as part of TriaLinQ. As an accelerator on its own prospective research, ASCO will continue to have access to CancerLinQ Discovery data, and CancerLinQ will support ASCO’s own prospective trials, such as the Targeted Agent and Profiling Utilization Registry (TAPUR) Study and the Patient-Centered Outcomes Research Institute (PCORI)-funded study of dosing strategies of oral CDK4/6 inhibitors in older adults living with metastatic breast cancer. 

As CancerLinQ moves into this stage, CancerLinQ LLC remains intact and the agreements between it and participating practices are not being changed.

“Through this process we have committed to the CancerLinQ members and ASCO to advance quality programs to the next-generation Certified, also consistent with the U.S. Department of Health and Human Services’ Oncology Care Model, the foundation to value-based care,” Dr. Elton said. “CancerLinQ can deliver value to clinical practices in an increasing number of areas. We will work collaboratively to assure this is fully realized and that it is the reference data source for oncology evidence generation.”