NCI’s pioneering clinical trial initiative—the Molecular Analysis for Therapy Choice (MATCH) Trial Network—heralded a new era for precision medicine studies. MATCH gave researchers a way to target therapies to people based on their specific genetic makeup. This approach helped make clinical trials more effective and gave people with cancer access to medications specifically tailored to their needs and situations.

The success of the original MATCH trial, which ended in 2023, spawned a collection of other precision medicine initiatives, including myeloMATCH, ComboMATCH, ImmunoMATCH, and Pediatric MATCH. Before embarking on these next-generation clinical trials, NCI looked for gaps and efficiencies in the overall process. NCI’s Center for Biomedical Informatics and Information Technology (CBIIT) then addressed these issues by developing a modular platform built on modern integration standards and state-of-the-art automation techniques.

In this blog, members of that CBIIT team share the lessons they learned in developing this platform. Their tips will be particularly useful if you’re automating a similar system for tracking and managing patients with cancer and their samples, or other research data.

Tip #1: Assemble a Multidisciplinary Team

We designed a CBIIT team consisting of members with specific skillsets to address the complexities inherent to precision medicine workflows, spanning patient entry, sample management, trial execution, and post-trial follow-up. If you’re undertaking similar work, you’ll want to consider having the following individuals on your team:

Software Engineers

Include people who can develop and implement the full range of platform components, such as intuitive user interfaces, robust sample tracking mechanisms, secure data management systems, and efficient backend processes. Select engineers who can develop software that allows for easy maintenance, evaluation, and minimal system downtime.

Informatics and Data Science Specialists

You’ll also need staff who can foster communication. These staff serve as critical intermediaries to connect external partners, analytical teams, software developers, and clinical protocol teams. Informaticists can establish structured methods for curating knowledge from current scientific literature, clinical reports, case studies, committee discussions, and interdisciplinary meetings.

User Support Specialists

If possible, include team members who can troubleshoot simple user problems. We implemented a dedicated help desk. These staff have the skills and know-how to swiftly resolve user issues, from basic troubleshooting and account recovery to software guidance.

Tip #2: Prioritize User-Centered Design

To address system usability, we conducted extensive user interface (UI) and user experience (UX) testing. Stakeholders—including clinicians, laboratory staff, administrative coordinators, and project support personnel—gave feedback throughout a design and development process. Using this approach, you can ensure your system’s functionalities align with real-world operational needs, enabling you to fix issues early in the process when changes are easier (and less costly) to make.

Tip #3: Implement Robust Data Security Protocols

If you’re working with sensitive data, you’ll want to take steps to protect personal information. With myeloMATCH and related trials, we’re managing genomic, clinical, and pharmaceutical data, so we priortized rigorous protocols to safeguard data security and integrity. Our approach included developing standardized protocols and software tools for consistent data representation and using anonymization strategies, such as de-identified or coded identifiers to protect personally identifiable information.

Tip #4: Leverage Automation and Standardized Terminologies to Enhance Efficiency

Are you modernizing an existing system? This is a good time to look for ways to make a workflow more efficient. This was a key priority for our clinical trial infrastructure, and we turned to automation, driven by standardized terminologies and coding automation techniques. Recognizing the critical need for rapid procurement, tracking, and analysis of molecular testing results, we incorporated widely adopted medical terminologies and coding standards (e.g., ICD-O-3) to streamline data integration, interoperability, and exchange across systems. We also developed advanced computational algorithms, using automated coding methods, to rapidly interpret patient genetic profiles and efficiently identify biomarkers. In all, we were able to trim the turnaround times for this work from months to days. At the same time, we significantly enhanced precision and efficiency by automating real-time notifications for sample status updates and workflows for integrating external laboratory data.

Tip #5: Offer Continuous, Accessible User Support

Consider putting a team in place to manage long-term support. Part of CBIIT’s ongoing support includes staffing a comprehensive help desk that’s operational 7 days per week, and which serves as the primary contact point for initial assistance. Our basic level-one help desk personnel use standardized procedures to efficiently resolve common issues, such as password resets and software troubleshooting, thus addressing most inquiries directly.

For complex issues requiring advanced support, we offer specialized assistance to all myeloMATCH stakeholders, including participants within the MDNet laboratory network and numerous oncology consortia, such as the Canadian Cancer Trials Group, SWOG Cancer Research Network, Eastern Cooperative Oncology Group, American College of Radiology Imaging Network, Alliance for Clinical Trials in Oncology, and Cancer Trials Support Unit.

With these groups, we often field questions on how to automate patient workflows, where to find user guides, and other frequently asked questions.