Dec 16, 2020
By Katherine H. Crawford, MA, PMP, ASCO Communications
Matching patients with cancer to a relevant clinical trial using their electronic health record (EHR) data; helping overworked clinicians and administrative staff at cancer centers report data to various registries that are aggregating data for various reasons; making clinical research more efficient and diverse by pulling data from a wider range of patients than traditional clinical trials—these are some of the promising developments powered by mCODE® (Minimal Common Oncology Data Elements) that are currently in the pilot phase.
All of these advancements are made possible through aggregating and sharing data among various EHRs. However, this type of data sharing has not come easily. There are dozens of different EHR systems used by hospitals, cancer centers, clinics, and oncology practices across the United States, and they don’t easily communicate with one another. Additionally, the great variation among clinicians in the way they input notes and comments and the way these notes aren’t captured in the same way as, for example, an item selected from a pulldown menu of options—means that much of this information is, essentially, locked away in an individual patient’s record, unable to be aggregated with others in a true learning health system.1
“Because most EHR systems were created to provide documentation for billing purposes and not for managing clinical and research data, outputs from them are little more than nonstandardized electronic paper,” wrote chair of the Association for Clinical Oncology and ASCO past president Monica M. Bertagnolli, MD, FACS, FASCO, et al. in a Journal of Clinical Oncology article published earlier this year. “This is especially true of the unstructured free text of the physician notes and diagnostic reports, which in oncology often contain the most valuable information, such as disease status (e.g., progressing v. responding to treatment), treatment rationale, and many other factors involved in the clinical judgment underlying treatment decisions.”
This “nonstandardized” format is a challenge for health care big data in general. As a 2018 NEJM Catalyst article stated, “'Big data in health care' refers to the abundant health data amassed from numerous sources including EHRs, medical imaging, genomic sequencing, payor records, pharmaceutical research, wearables, and medical devices, to name a few. Three characteristics distinguish it from traditional electronic medical and human health data used for decision-making: It is available in extraordinarily high volume; it moves at high velocity and spans the health industry’s massive digital universe; and, because it derives from many sources, it is highly variable in structure and nature. This is known as the 3Vs of Big Data."
In the field of cancer care, a group of organizations has come together to develop a solution that will reduce variability among the many codes used to denote care received at an oncology practice: mCODE.
The mCODE (Minimal Common Oncology Data Elements) initiative is a collaboration between ASCO, the MITRE Corporation, the Alliance for Clinical Trials in Oncology Foundation, the American Society for Radiation Oncology, and the Society for Surgical Oncology that has developed a core set of structured data elements to facilitate cancer data interoperability.
The mCODE elements, which are available for use free of charge, are designed using standard, widely available medical terminologies to enable searchability. In practice, this means that when oncology care team members search different EHRs that all use mCODE, they should be able to retrieve similar results: the data can be found, accessed, and made available.
The first public version of mCODE was released during the ASCO Annual Meeting in 2019. In March 2020, mCODE Version 1.0 was formally approved by Health Level 7 International (HL7) as a Standard for Trial Use. mCODE contains 90 data elements that are grouped into six main areas: patient, disease, genomics, labs/vitals, treatment, and outcomes. Some of the elements are being used in two of ASCO’s products, CancerLinQ and the ASCO Survey on COVID-19 in Oncology Registry.
A Collaborative Effort
“The mCODE initiative has been a collaborative venture from its inception,” said CancerLinQ medical director Robert S. Miller, MD, FACP, FASCO, FAMIA. “If we are truly going to increase interoperability, we need to come together—oncologists, health care systems, EHR vendors, and data experts—and talk through the challenges we face and potential solutions toward generating a common list of data elements to support patient care and other critical use cases.”
In early 2020, the mCODE Council was established with 20 members from academic, community practice, government, industry, and nonprofit organizations. The Council provides thought leadership, advises on new and amended data elements, suggests and advises on use cases, and promotes the initiative. Organizations interested in joining the Council can fill out a brief application to be considered for membership.
Survey on Prioritizing Use Case Topics
The mCODE Council met in July 2020, and in advance of that meeting they completed a brief survey about prioritizing use case topics—those unique examples in which a specific type of information regarding patients and their health is shared—to help inform mCODE Executive Committee discussions on mCODE’s strategic direction. In the future, the committee plans to formally identify a set of priorities for future expansion of the data dictionary and provide a framework to guide community efforts. The survey, which will help shape these actions, revealed that:
- Clinical trial data capture is a priority topic for use cases; additional topics of interest included cancer registry reporting and oncology clinical pathways.
- Data items for regulatory reporting (e.g. the Food and Drug Administration’s Standardizing and Evaluating Risk Evaluation and Mitigation Strategies [REMS]) is another priority area for use cases.
- Data entry burden and impact on clinician workflow are key barriers to mCODE’s adoption.
- Those with firsthand experience of dealing with challenges in data interoperability—whether from a clinician, patient, or technology perspective—are very motivated to work toward creating a learning health system.
This latter point further emphasizes the important role that mCODE can play in furthering the body of oncology knowledge. As Dr. Bertagnolli stated to attendees of the 2019 ASCO Annual Meeting in her presidential address, “By sharing your decisions and results using the common mCODE language, you will make sure that translation errors do not limit the quality of the information that you and your patients contribute to the learning environment.”
While the mCODE Executive Committee sets the initiative’s future course, the current iteration of elements are available for use free of charge, and they are in use both domestically and outside the United States. One example of their use abroad is from the National Health Research Institutes in Taiwan, who referenced mCODE elements during their planning to establish a common data model for aggregating data elements from EHRs at all of the sites. This work to develop a common set of oncology data and aid interoperability is outlined in an abstract presented at the ASCO20 Virtual Scientific Program.
ICAREdata Study Pilot
Here in the United States, a new method for collecting data from clinical trials is being piloted.
The ICAREdata Study is piloting the use of mCODE to aid in collecting high-quality real-world data to enable clinical oncology research. The study staff are collaborating with clinicians involved in clinical trials to collect important outcome data that, at present, are not well represented in EHRs and then share this data with clinical sites and mCODE. The hope is that using a patient’s EHR as a source of study data will dramatically improve clinical trial operational efficiency and reduce costs.
The pilot is now inits second stage, wherein the Alliance for Clinical Trials in Oncology and MITRE are working with clinical trial staff to demonstrate that real-world data from an EHR based on mCODE can support equivalent clinical trial assessments. Data collected from the EHR of study participants include those defining primary study endpoints, including disease treatment response and adverse events. The first Alliance trials where these data elements are being assessed are:
- A071701 – Genomically Guided Treatment Trial in Brain Metastases
- A021703 – SOLARIS: Randomized Double-blind Phase III Trial of Vitamin D Supplementation in Patients with Previously Untreated Metastatic Colorectal Cancer
- A031902 – CASPAR: Phase III Trial of Enzalutamide and Rucaparib as a Novel Therapy in First-Line Metastatic Castration-Resistant Prostate Cancer
The Alliance and MITRE are seeking other clinical trials sites to join the five that are currently participating in the pilot: Brigham and Women’s Hospital, Dana Farber Cancer Institute, Massachusetts General Hospital, the Mayo Clinic, and The University of Pennsylvania Health System.
The mCODE initiative is continuing to develop and evolve. The mCODE Technical Review Group, which reports to the mCODE Executive Committee, is responsible for maintaining the data dictionary and reviewing new use cases that are proposed, as well as reviewing comments and suggestions made by the broader community to identify challenges and opportunities to expand mCODE as the standard comes into wider use. This is so important because, as David A. Chambers, DPhil, et al. stated in a 2019 ASCO Educational Book article about challenges and solutions in using health care big data to inform clinical care, “the use of multiple sources of big data has the promise of improving knowledge and providing more accurate data for clinicians and policy decision makers.” And if that data can be made more useful and user-friendly to clinicians, researchers, and others in the oncology community, patients will reap the benefits—and ultimately improve their quality of care.
In early 2020, the mCODE Council was established with 20 members from academic, community practice, government, industry, and nonprofit organizations. The Council provides thought leadership, advises on new and amended data elements, suggests and advises on use cases, and promotes the initiative.
In addition to the collaborators that developed mCODE—ASCO, the MITRE Corporation, the Alliance for Clinical Trials in Oncology Foundation, the American Society for Radiation Oncology, and the Society for Surgical Oncology—the organizations with representatives on the Council are:
- American Cancer Society Cancer Action Network
- American Society of Hematology
- Clinical Data Interchange Standards Consortium (CDISC)
- Center for International Blood and Marrow Transplant Research (CIBMTR)
- College of American Pathologists
- DeepThink Health
- Friends of Cancer Research
- Intermountain Healthcare
- Mission Cancer + Blood
- Mount Sinai Health System
- National Cancer Institute
- Oncology Nursing Society
- Rush Health
- University of California - San Francisco
- University of Pennsylvania
Organizations interested in joining the Council can fill out a brief application to be considered for membership.
- Osterman T, Terry M, Miller RS. Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative. JCO Clin Cancer Inform. 2020;4:993-1001.