MITRE Corp., is leading development of a common data model in oncology called mCODE (minimal Common Oncology Data Elements). This initiative is identifying cancer data elements that would be essential for analyzing treatment across electronic health records (EHRs) and cancer practices to improve quality and care coordination.
To put this effort into practice, the mCODE team is working with the Alliance for Clinical Trials in Oncology on a project called Integrating Clinical Trials and Real-World Endpoints (ICARE), which aims to demonstrate that mCODE-based real-world data can drive more efficient clinical research by incorporating the data for a broader population of patients while maintaining the same quality as traditional clinical trials. The goal of the ICARE data study is to enable clinical oncology research by gathering real-world data from patient records.
Speaking at a July 14 National Cancer Policy Forum virtual meeting on using digital health applications in oncology, Monica M. Bertagnolli, M.D., Group Chair of the Alliance for Clinical Trials in Oncology and Chief of the Division of Surgical Oncology at Dana-Farber Cancer Institute/Brigham and Women’s Hospital, described ICARE.
She said the Alliance is working with MITRE, Epic and the National Cancer Institute, funded partially by the Rising Tide Foundation. The use case “looks at case report forms from a variety of clinical trials and asks a question: How well does mCODE in its first version cover the data that is required to complete case report forms for an Alliance clinical trial?”
The study is initially focusing on disease progression and toxicity and will expand to cover a broader set of data across multiple clinical trials moving forward.
First they mapped the Alliance case report form to mCODE. Then they found out how much data could be converted to mCODE and extracted it from the EHR. Then working with partners at MITRE, they developed an implementation priority to decide which elements needed to be pulled from the EHR.
“The MITRE Group has a team working with us called STEAM, which stands for Scalable Technology for EHR Adoption of mCODE. They partnered with Epic to develop an implementation that allowed us to extract clinical trial data elements from the EHR of the Epic instance,” Bertagnolli explained. “They developed a playbook that described exactly how these data elements could be configured and coded so they can be extracted. Epic provided implementation allowing the data to be pulled from the EHR via an application programming interface into the Alliance database.”
“This is an example of a use case where we took the mCODE data elements, decided which would suit our needs, developed a method to obtain those data structures from the EHR and are now launching them in our clinical trials,” she added. “Finally, we found out that everything we needed was not in mCODE. So what did we do? We provided information about these gaps back to the CodeX community, where we will work on new iterations of mCODE that will more fully serve the needs of the cancer clinical trials community. (CodeX is an HL7 FHIR accelerator leveraging the mCODE FHIR Implementation Guide.)
There are other CodeX use cases under way, Bertagnolli said. There is a clinical trials patient- matching use case led by the American Cancer Society Cancer Action Network, and there are registry reporting use cases under way, a radiation oncology care coordination use case, clinical care pathways, and a prior authorization use case. All of these are under execution by CodeX, and all these organizations are participating in its community of practice.
MITRE has developed Compass, an mCODE-based FHIR implementation and SMART-on-FHIR application. Intermountain Healthcare is testing Compass to extract mCODE data elements from EHRs to create reports for providers and patients that can be used to make shared treatment decisions based on which treatments worked for similar cohorts of patients.
MITRE is also developing Camino, a SMART-on-FHIR application that provides computable standards-based clinical pathways. Navigation through the protocol is driven by mCODE data elements accessible from the EHR. Camino allows the oncologist to see both where the patient is on the pathway and recommended treatment options.
