Yu-Cherng Chang1, Radka Stoyanova2,3, Sofia Danilova4, Juan A Pretell-Mazzini3,5, Darcy A Kerr6,7, Breelyn A Wilky8, Ty Subhawong9,10. 1. Miller School of Medicine, University of Miami, Miami, FL, USA. 2. Department of Radiation Oncology, University of Miami Miller School of Medicine, Miami, FL, USA. 3. Sylvester Comprehensive Cancer Center, Miami, FL, USA. 4. Department of Radiology, University of Miami Miller School of Medicine, Miami, FL, USA. 5. Department of Orthopedics, University of Miami Miller School of Medicine, Miami, FL, USA. 6. Department of Pathology and Laboratory Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA. 7. Geisel School of Medicine at Dartmouth, Hanover, NH, USA. 8. Department of Medicine-Medical Oncology, University of Colorado Hospital, Aurora, CO, USA. 9. Sylvester Comprehensive Cancer Center, Miami, FL, USA. tsubhawong@med.miami.edu. 10. Department of Radiology, University of Miami Miller School of Medicine, Miami, FL, USA. tsubhawong@med.miami.edu.
Abstract
OBJECTIVE: Denosumab is an established targeted systemic therapy for treatment of giant cell tumor of bone (GCTB). We sought to determine whether treatment response could be quantified from radiomics analysis of radiographs taken longitudinally during treatment. MATERIALS AND METHODS: Pre- and post-treatment radiographs of 10 GCTB tumors from 10 patients demonstrating histologic response after treatment with denosumab were analyzed. Intensity- and texture-based radiomics features for each manually segmented tumor were calculated. Radiomics features were compared pre- and post-treatment in tumors. RESULTS: Mean intensity (p = 0.033) significantly increased while skewness (p = 0.028) significantly decreased after treatment. Post-treatment increases in fractal dimensions (p = 0.057) and abundance (p = 0.065) approached significance. A potential linear correlation in mean (p = 0.005; ΔMean = 0.022 * duration - 0.026) with treatment duration was observed. CONCLUSION: Radiomics analysis of plain radiographs quantifies time-dependent matrix mineralization and trabecular reconstitution that mark positive response of giant cell tumors of bone to denosumab.
OBJECTIVE:Denosumab is an established targeted systemic therapy for treatment of giant cell tumor of bone (GCTB). We sought to determine whether treatment response could be quantified from radiomics analysis of radiographs taken longitudinally during treatment. MATERIALS AND METHODS: Pre- and post-treatment radiographs of 10 GCTB tumors from 10 patients demonstrating histologic response after treatment with denosumab were analyzed. Intensity- and texture-based radiomics features for each manually segmented tumor were calculated. Radiomics features were compared pre- and post-treatment in tumors. RESULTS: Mean intensity (p = 0.033) significantly increased while skewness (p = 0.028) significantly decreased after treatment. Post-treatment increases in fractal dimensions (p = 0.057) and abundance (p = 0.065) approached significance. A potential linear correlation in mean (p = 0.005; ΔMean = 0.022 * duration - 0.026) with treatment duration was observed. CONCLUSION: Radiomics analysis of plain radiographs quantifies time-dependent matrix mineralization and trabecular reconstitution that mark positive response of giant cell tumors of bone to denosumab.