Author: Gregory T. Armstrong, James E. Bates, Kristy K. Brock, Laurence Edward Court, Matt Ehrhardt, Danielle Friedman, Aashish C. Gupta, Donald Hancock, Rebecca M. Howell, Cindy Im, Tera S Jones, Choonsik Lee, Wendy Leisenring, Taylor Meyers, Lindsay Morton, Chaya Moskowitz, Joe Neglia, Vikki Nolan, Caleb O'Connor, Kevin C. Oeffinger, Constance A. Owens, Arnold C. Paulino, Chelsea C. Pinnix, Sander Roberti, Cecile Ronckers, Susan A. Smith, Kumar Srivastava, Lucie Turcotte 👨🔬
Affiliation: Department of Medicine, Duke University School of Medicine, Department of Epidemiology and Cancer Control, St. Jude Children’s Research Hospital, The University of Texas MD Anderson Cancer Center, Department of Oncology, St. Jude Children’s Research Hospital, Department of Pediatrics, Memorial Sloan Kettering Cancer Center, Division of Pediatric Epidemiology and Clinical Research, University of Minnesota, Division of Childhood Cancer Epidemiology, University Medicine at Johannes Gutenberg University Mainz, Department of Radiation Physics, The University of Texas MD Anderson Cancer Center, Department of Pediatrics, University of Minnesota, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Department of Biostatistics, St. Jude Children’s Research Hospital, Clinical Research Division, Fred Hutchinson Cancer Center, Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Department of Radiation Oncology and Winship Cancer Institute, Emory University, The University of Texas MD Anderson Cancer Center UTHealth Houston Graduate School of Biomedical Sciences 🌍
Purpose: To (1) develop and validate a novel anatomically realistic pediatric/adolescent population-based breast model, (2) incorporate model into an age-scalable female reference phantom, and (3) demonstrate feasibility for pre-CT era radiotherapy dose reconstruction for female childhood/adolescent cancer survivors.
Methods: Breast contours from 79 female Hodgkin lymphoma patients’ (12–21 years) CTs (one-reference, 70-training, eight-testing) were collected. Principal component analysis statistical shape modeling (PCA-SSM) was performed on training contours to capture population deformations. A mean breast model and two alternative shapes were integrated into an age-scalable phantom. Geometric accuracy was assessed using Dice similarity coefficient (DSC), distance-to-agreement (DTA), and Hausdorff distance (HD). Dosimetric accuracy was evaluated by comparing reconstructed and ground-truth breast and breast quadrant dose-volume metrics from mantle-field radiotherapy for test patients’ and age-matched phantoms. Radiotherapy was reconstructed for 663 female survivors in the Childhood Cancer Survivor Cohort (CCSS) diagnosed 1970-1999 at 31 institutions aged 12-20 years.
Results: DSC(min-max) was 0.86(0.85-0.88), DTA was 1.46mm(1.21-2.04mm), and 95th HD was 2.44mm(1.98-3.32mm). Absolute differences for mean breast doses (normalized to prescription) between ground truth and mean model (alternative#1, alternative#2) were 2.63%(3.57%, 2.22%). For lower quadrants, differences were within 10%, while upper-inner and upper-outer quadrant differences were 16.12%(20.49%, 10.70%) and 10.06%(9.65%, 10.53%), respectively. Dose-volume metric differences (V5–V30) were typically within 5%, with some cases reaching 13%. Among 663 CCSS survivors, reconstructed mean(min-max) breast doses, V5, and V20 were 8.15Gy(0.05Gy-44.37Gy), 35.75%(0-100%), and 14.25%(0-100%), respectively.
Conclusion: A population-based pediatric/adolescent breast model was developed, validated, and used to reconstruct breast doses for a subset of CCSS participants, a cohort that previously reported subsequent breast cancer (SBC) dose-response models based on prescribed chest doses (breast dose surrogate). This work makes it possible to complete breast dosimetry for the CCSS and develop novel breast dose-volume-based SBC response models to define dose-volume constraints for newly diagnosed girls/adolescents and guide survivorship care.