Tumor Size Is Not Everything: Advancing Radiomics as a Precision Medicine Biomarker in Oncology Drug Development and Clinical Care
In contemporary clinical oncology practice and drug development, the methods for evaluating tumor response are on the cusp of a revolution. Since the World Health Organization (WHO) proposed tumor response classification criteria for assessing the effectiveness of anti-cancer drugs in 1981, this field has undergone several improvements. Notably, the Response Evaluation Criteria in Solid Tumors (RECIST) working group, established in 1995, is well-known. This group, in collaboration with the Canadian National Cancer Institute, the U.S. National Cancer Institute, and the European Organization for Research and Treatment of Cancer, developed evidence-based recommendations based on extensive case data, leading to the release of RECIST 1.0 and 1.1 versions, which have played a critical role in determining objective response rates and other imaging endpoints.
However, with deeper understanding of tumor biology and continuous evolution of diagnostic and therapeutic strategies, traditional assessment methods relying on size and quantity changes have shown their limitations, necessitating new methods to fill the gap. To this end, the RECIST working group organized a multidisciplinary workshop in May 2022, focusing on the potential applications of radiomics in oncology drug development and clinical trials, particularly in response evaluation.
Radiomics, as a science that extracts quantifiable data from routine medical images, provides a new dimension for the assessment of tumor characteristics and treatment response. Traditional imaging assessments are primarily based on the interpretation by clinicians and simple two-dimensional measurements, whereas radiomics extracts a large number of features from images and analyzes this data using bioinformatics tools and artificial intelligence technologies, thereby revealing the molecular, genomic, structural, and microenvironmental characteristics of tumors. This not only provides new perspectives for understanding tumor biology but also supports personalized medical decision-making.
The workshop, chaired by Erica C. Nakajima, MD, and published in the journal “JCO Precision Oncology,” aims to review the applications of radiomics in clinical oncology practice and highlight the necessary work to integrate radiomics into clinical trials. The article emphasizes the potential of radiomics as a biomarker for precision medicine, along with the challenges and future steps.
The application range of radiomics is extensive, from refined cancer screening to early evaluation of tumor treatment response and even predicting tumor molecular signatures. However, despite its significant potential, the application of radiomics in routine clinical practice remains limited. The paper highlights a series of important research findings demonstrating the ability of radiomics in predicting treatment response and improving the design of clinical trials.
To translate the advancements in radiomics into practical outcomes in tumor treatment, interdisciplinary collaboration is required, involving radiologists, clinical oncologists, bioinformaticians, and computer scientists, to collectively overcome technical and methodological barriers. Furthermore, establishing the link between radiomic features and tumor biomarkers will be a crucial step towards achieving clinical application. Future clinical trial designs need to integrate and test radiomics as companion biomarkers.
This paper not only outlines the potential applications and challenges of radiomics in the field of oncology but also proposes clear directions and recommendations for further exploration and application. It aims to accelerate the integration and application of radiomics as a precision medicine tool through the design and execution of clinical trials.