Jennifer Yin Yee Kwan1, Jie Su2, Shao Hui Huang1, Laleh S Ghoraie3, Wei Xu2, Biu Chan1, Kenneth W Yip3, Meredith Giuliani1, Andrew Bayley1, John Kim1, Andrew J Hope1, Jolie Ringash1, John Cho1, Andrea McNiven1, Aaron Hansen4, David Goldstein5, John R de Almeida5, Hugo J Aerts6, John N Waldron1, Benjamin Haibe-Kains7, Brian O'Sullivan1, Scott V Bratman1, Fei-Fei Liu8. 1. Department of Radiation Oncology, Princess Margaret Cancer Centre/University of Toronto, Toronto, Ontario, Canada. 2. Department of Biostatistics, University Health Network, Toronto, Ontario, Canada. 3. Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada. 4. Division of Medical Oncology, University of Toronto, Toronto, Ontario, Canada. 5. Department of Otolaryngology-Head and Neck Surgery and Surgical Oncology, University of Toronto, Toronto, Ontario, Canada. 6. Department of Radiation Oncology, Dana-Farber Cancer Institute, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts; Department of Radiology, Dana-Farber Cancer Institute, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts. 7. Princess Margaret Cancer Centre, University Health Network, Toronto, Ontario, Canada; Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; Department of Computer Science, University of Toronto, Toronto, Ontario, Canada; Ontario Institute of Cancer Research, Toronto, Ontario, Canada. 8. Department of Radiation Oncology, Princess Margaret Cancer Centre/University of Toronto, Toronto, Ontario, Canada; Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada. Electronic address: Fei-Fei.Liu@rmp.uhn.on.ca.
Abstract
PURPOSE: Distant metastasis (DM) is the main cause of death for patients with human papillomavirus (HPV)-related oropharyngeal cancers (OPCs); yet, there are few reliable predictors of DM in this disease. The role of quantitative imaging (ie, radiomic) analysis was examined to determine whether there are primary tumor features discernible on imaging studies that are associated with a higher risk of DM developing. METHODS AND MATERIALS: Radiation therapy planning computed tomography scans were retrieved for all nonmetastatic p16-positive OPC patients treated with radiation therapy or chemoradiation therapy at a single institution between 2005 and 2010. Radiomic biomarkers were derived from each gross tumor volume. The biomarkers included 4 representative radiomic features from tumor first-order statistics, shape, texture, and wavelet groups, as well as a combined 4-feature signature. Univariable Cox proportional hazards models for DM risk were identified. The discriminative performance of prognostic univariable and multivariable models was compared using the concordance index (C-index). Subgroup analyses were performed. RESULTS: There were 300 HPV-related OPC patients who were eligible for the analysis. A total of 36 DM events occurred within a median follow-up period of 5 years. On univariable analysis, top results included the 4 representative radiomic features (C-index, 0.670-0.686; P < .001), the radiomic signature (C-index, 0.670; P < .001), tumor stage (C-index, 0.633; P < .001), tumor diameter (C-index, 0.653; P < .001), and tumor volume (C-index, 0.674; P < .001), which demonstrated moderate discrimination of DM risk. Combined clinical-radiomic models yielded significantly improved performance (C-index, 0.701-0.714; P < .05). In subgroup analyses, the radiomic biomarkers consistently stratified patients for DM risk, particularly for those cohorts with greater risks (C-index, 0.663-0.796), such as patients with stage III disease. CONCLUSIONS: Radiomic biomarkers appear to classify DM risk for patients with nonmetastatic HPV-related OPC. Radiomic biomarkers could be used either alone or with other clinical characteristics in the assignment of DM risk in future HPV-related OPC clinical trials.
PURPOSE: Distant metastasis (DM) is the main cause of death for patients with human papillomavirus (HPV)-related oropharyngeal cancers (OPCs); yet, there are few reliable predictors of DM in this disease. The role of quantitative imaging (ie, radiomic) analysis was examined to determine whether there are primary tumor features discernible on imaging studies that are associated with a higher risk of DM developing. METHODS AND MATERIALS: Radiation therapy planning computed tomography scans were retrieved for all nonmetastatic p16-positive OPCpatients treated with radiation therapy or chemoradiation therapy at a single institution between 2005 and 2010. Radiomic biomarkers were derived from each gross tumor volume. The biomarkers included 4 representative radiomic features from tumor first-order statistics, shape, texture, and wavelet groups, as well as a combined 4-feature signature. Univariable Cox proportional hazards models for DM risk were identified. The discriminative performance of prognostic univariable and multivariable models was compared using the concordance index (C-index). Subgroup analyses were performed. RESULTS: There were 300 HPV-related OPCpatients who were eligible for the analysis. A total of 36 DM events occurred within a median follow-up period of 5 years. On univariable analysis, top results included the 4 representative radiomic features (C-index, 0.670-0.686; P < .001), the radiomic signature (C-index, 0.670; P < .001), tumor stage (C-index, 0.633; P < .001), tumor diameter (C-index, 0.653; P < .001), and tumor volume (C-index, 0.674; P < .001), which demonstrated moderate discrimination of DM risk. Combined clinical-radiomic models yielded significantly improved performance (C-index, 0.701-0.714; P < .05). In subgroup analyses, the radiomic biomarkers consistently stratified patients for DM risk, particularly for those cohorts with greater risks (C-index, 0.663-0.796), such as patients with stage III disease. CONCLUSIONS: Radiomic biomarkers appear to classify DM risk for patients with nonmetastatic HPV-related OPC. Radiomic biomarkers could be used either alone or with other clinical characteristics in the assignment of DM risk in future HPV-related OPC clinical trials.
Authors: Zezhong Ye; Jack M Qian; Ahmed Hosny; Roman Zeleznik; Deborah Plana; Jirapat Likitlersuang; Zhongyi Zhang; Raymond H Mak; Hugo J W L Aerts; Benjamin H Kann Journal: Radiol Artif Intell Date: 2022-05-04