Standards for Quantitative Imaging Biomarkers to Advance Research and Outcomes as part of the Cancer Moonshot
By Daniel Sullivan, Duke University Medical Center; Roderic Pettigrew, National Institute of Biomedical Imaging and Bioengineering; Richard Cavanagh, National Institute of Standards and Technology; and Shadi Mamaghani, National Institute of Biomedical Imaging and Bioengineering
Among the Cancer Moonshot initiative goals are unleashing the power of data and enhancing data sharing. Efforts to facilitate and promote data sharing are crucial to both researchers and clinicians. With improved sharing of datasets, researchers can speed scientific discovery and the development of effective treatments and diagnostics; care providers can improve diagnosis, treatment, and ultimately health outcomes. Medical imaging plays a central role in cancer treatment and research: each year in the United States, about 23 million computed tomography (CT) scans and 1.6 million positron emission tomography (PET) scans are performed. These scans are needed to diagnose cancer, monitor the effects of treatments, and conduct research on cancer and cancer-related therapies. Scans can also assist in understanding the genomic expression of cancer and its response to gene-based therapies.
1. The Need for Standards
To obtain the greatest value from cancer imaging the scans must be compared across time for a given patient, and across patients and research institutions. This is not a trivial undertaking: image-based assessments of treatment responses can be highly variable, depending on the make and model of the imaging equipment used and how the images were acquired, processed, and the guidelines by which they were interpreted. If images are not uniformly acquired, doctors and researchers may not be able to compare the results of one study with another in the same patient, between groups of patients in a clinical trial, or between studies or institutions. For example, if a cancer patient has a PET scan done at one hospital using one type of scanner and procedures, and two months after treatment has a PET scan on a different scanner or with different procedures, measurements from the two scans might not be comparable. That is, it may be difficult to determine whether any apparent changes between the two studies are due to real changes in the cancer or due to technical differences between the scans.
2. QIBA: An Alliance to Establish Imaging Standards
The Radiological Society of North America (RSNA) has established the Quantitative Imaging Biomarkers Alliance (QIBA) to address this problem. QIBA aims to improve the value and practicality of quantitative imaging biomarkers by reducing variability across devices and time. Over 1000 experts across science, engineering and medicine support the activities to develop standards and advance healthcare using quantitative imaging biomarkers in clinical trials and practice. QIBA has adapted standard terminology and methods traditionally used to describe, evaluate and validate in vitro laboratory assays, and has applied these to quantitative imaging biomarkers. Standardizing the use of quantitative imaging biomarkers can reduce the variance across hardware and software platforms used in different research and treatment centers, and make it more reliable to compare the results of one imaging study with another. A systematic, consensus-driven approach generates “QIBA Profiles” which define universal standards of operation. These Profiles specifically establish comprehensive, systems-engineering, technical standards for image acquisition and processing.
3. QIBA Standards Established for Two Major Imaging Procedures
Two standardized Profiles relevant for acquiring and analyzing images used in cancer studies and treatments are complete; these describe methods for obtaining: 1) accurate and reproducible 18F-fluorodeoxyglucose (FDG) PET/CT measurements; and 2) CT tumor volume measurements. Of note is that a growing number of medical practices, imaging vendors, research organizations, and other entities are announcing their commitment to endorse or adopt these QIBA Profiles to improve the reliability of cancer imaging studies. These statements of support mean that more imaging equipment will conform to QIBA standards and more imaging centers will adopt standardized practices for conducting imaging studies.
Use of these QIBA Profiles for standardized quantitative imaging will contribute significantly to improvements in the quality of cancer care, and to the development of more effective therapeutics in oncology. It will help achieve the goal of high-value imaging recently articulated by the consortium of twelve Federal agencies participating in the Interagency Working Group on Medical Imaging.
Daniel Sullivan, M.D. is Professor Emeritus in the Department of Radiology at the Duke University Medical Center and External Relations Liaison, Quantitative Imaging Biomarkers Alliance.
Roderic Pettigrew, Ph.D., M.D., is Director of the National Institute of Biomedical Imaging and Bioengineering at the National Institutes of Health.
Richard Cavanagh, Ph.D. is Director of the Special Programs Office at the National Institute of Standards and Technology.
Shadi Mamaghani, Ph.D. is Senior Scientific Advisor, Office of the Director, National Institute of Biomedical Imaging and Bioengineering.