Maria Thor1, Aditya Apte2, Rabia Haq2, Aditi Iyer2, Eve LoCastro2, Joseph O Deasy2. 1. Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York. Electronic address: thorm@mskcc.org. 2. Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York.
Abstract
PURPOSE: Contouring inconsistencies are known but understudied in clinical radiation therapy trials. We applied auto-contouring to the Radiation Therapy Oncology Group (RTOG) 0617 dose escalation trial data. We hypothesized that the trial heart doses were higher than reported due to inconsistent and insufficient heart segmentation. We tested our hypothesis by comparing doses between deep-learning (DL) segmented hearts and trial hearts. METHODS AND MATERIALS: The RTOG 0617 data were downloaded from The Cancer Imaging Archive; the 442 patients with trial hearts and dose distributions were included. All hearts were resegmented using our DL pipeline and quality assured to meet the requirements for clinical implementation. Dose (V5%, V30%, and mean heart dose) was compared between the 2 sets of hearts (Wilcoxon signed-rank test). Each dose metric was associated with overall survival (Cox proportional hazards). Lastly, 18 volume similarity metrics were assessed for the hearts and correlated with |DoseDL - DoseRTOG0617| (linear regression; significance: P ≤ .0028; corrected for 18 tests). RESULTS: Dose metrics were significantly higher for DL hearts compared with trial hearts (eg, mean heart dose: 15 Gy vs 12 Gy; P = 5.8E-16). All 3 DL heart dose metrics were stronger overall survival predictors than those of the trial hearts (median, P = 2.8E-5 vs 2.0E-4). Thirteen similarity metrics explained |DoseDL - DoseRTOG0617|; the axial distance between the 2 centers of mass was the strongest predictor (CENTAxial; median, R2 = 0.47; P = 6.1E-62). CENTAxial agreed with the qualitatively identified inconsistencies in the superior direction. The trial's qualitative heart contouring score was not correlated with |DoseDL - DoseRTOG0617| (median, R2 = 0.01; P = .02) or with any of the similarity metrics (median, Rs = 0.13 [range, -0.22 to 0.31]). CONCLUSIONS: Using a coherent heart definition, as enabled through our open-source DL algorithm, the trial heart doses in RTOG 0617 were found to be significantly higher than previously reported, which may have led to an even more rapid mortality accumulation. Auto-segmentation is likely to reduce contouring and dose inconsistencies and increase the quality of clinical RT trials.
PURPOSE: Contouring inconsistencies are known but understudied in clinical radiation therapy trials. We applied auto-contouring to the Radiation Therapy Oncology Group (RTOG) 0617 dose escalation trial data. We hypothesized that the trial heart doses were higher than reported due to inconsistent and insufficient heart segmentation. We tested our hypothesis by comparing doses between deep-learning (DL) segmented hearts and trial hearts. METHODS AND MATERIALS: The RTOG 0617 data were downloaded from The Cancer Imaging Archive; the 442 patients with trial hearts and dose distributions were included. All hearts were resegmented using our DL pipeline and quality assured to meet the requirements for clinical implementation. Dose (V5%, V30%, and mean heart dose) was compared between the 2 sets of hearts (Wilcoxon signed-rank test). Each dose metric was associated with overall survival (Cox proportional hazards). Lastly, 18 volume similarity metrics were assessed for the hearts and correlated with |DoseDL - DoseRTOG0617| (linear regression; significance: P ≤ .0028; corrected for 18 tests). RESULTS: Dose metrics were significantly higher for DL hearts compared with trial hearts (eg, mean heart dose: 15 Gy vs 12 Gy; P = 5.8E-16). All 3 DL heart dose metrics were stronger overall survival predictors than those of the trial hearts (median, P = 2.8E-5 vs 2.0E-4). Thirteen similarity metrics explained |DoseDL - DoseRTOG0617|; the axial distance between the 2 centers of mass was the strongest predictor (CENTAxial; median, R2 = 0.47; P = 6.1E-62). CENTAxial agreed with the qualitatively identified inconsistencies in the superior direction. The trial's qualitative heart contouring score was not correlated with |DoseDL - DoseRTOG0617| (median, R2 = 0.01; P = .02) or with any of the similarity metrics (median, Rs = 0.13 [range, -0.22 to 0.31]). CONCLUSIONS: Using a coherent heart definition, as enabled through our open-source DL algorithm, the trial heart doses in RTOG 0617 were found to be significantly higher than previously reported, which may have led to an even more rapid mortality accumulation. Auto-segmentation is likely to reduce contouring and dose inconsistencies and increase the quality of clinical RT trials.
Authors: Michael G Jameson; Lois C Holloway; Philip J Vial; Shalini K Vinod; Peter E Metcalfe Journal: J Med Imaging Radiat Oncol Date: 2010-10 Impact factor: 1.735
Authors: Maria Thor; Margaret Montovano; Alexandra Hotca; Leo Luo; Andrew Jackson; Abraham J Wu; Joseph O Deasy; Andreas Rimner Journal: Radiother Oncol Date: 2019-10-12 Impact factor: 6.280
Authors: Barbara Stam; Heike Peulen; Matthias Guckenberger; Frederick Mantel; Andrew Hope; Maria Werner-Wasik; Jose Belderbos; Inga Grills; Nicolette O'Connell; Jan-Jakob Sonke Journal: Radiother Oncol Date: 2017-05-02 Impact factor: 6.280
Authors: Ward van Rooij; Max Dahele; Hugo Ribeiro Brandao; Alexander R Delaney; Berend J Slotman; Wilko F Verbakel Journal: Int J Radiat Oncol Biol Phys Date: 2019-03-02 Impact factor: 7.038
Authors: Jeffrey D Bradley; Chen Hu; Ritsuko R Komaki; Gregory A Masters; George R Blumenschein; Steven E Schild; Jeffrey A Bogart; Kenneth M Forster; Anthony M Magliocco; Vivek S Kavadi; Samir Narayan; Puneeth Iyengar; Clifford G Robinson; Raymond B Wynn; Christopher D Koprowski; Michael R Olson; Joanne Meng; Rebecca Paulus; Walter J Curran; Hak Choy Journal: J Clin Oncol Date: 2019-12-16 Impact factor: 44.544
Authors: Jeffrey D Bradley; Rebecca Paulus; Ritsuko Komaki; Gregory Masters; George Blumenschein; Steven Schild; Jeffrey Bogart; Chen Hu; Kenneth Forster; Anthony Magliocco; Vivek Kavadi; Yolanda I Garces; Samir Narayan; Puneeth Iyengar; Cliff Robinson; Raymond B Wynn; Christopher Koprowski; Joanne Meng; Jonathan Beitler; Rakesh Gaur; Walter Curran; Hak Choy Journal: Lancet Oncol Date: 2015-01-16 Impact factor: 41.316
Authors: Asma Amjad; Jiaofeng Xu; Dan Thill; Colleen Lawton; William Hall; Musaddiq J Awan; Monica Shukla; Beth A Erickson; X Allen Li Journal: Med Phys Date: 2022-02-07 Impact factor: 4.071
Authors: Maria Thor; Aditi Iyer; Jue Jiang; Aditya Apte; Harini Veeraraghavan; Natasha B Allgood; Jennifer A Kouri; Ying Zhou; Eve LoCastro; Sharif Elguindi; Linda Hong; Margie Hunt; Laura Cerviño; Michalis Aristophanous; Masoud Zarepisheh; Joseph O Deasy Journal: Phys Imaging Radiat Oncol Date: 2021-07-28
Authors: Gerard M Walls; Valentina Giacometti; Aditya Apte; Maria Thor; Conor McCann; Gerard G Hanna; John O'Connor; Joseph O Deasy; Alan R Hounsell; Karl T Butterworth; Aidan J Cole; Suneil Jain; Conor K McGarry Journal: Phys Imaging Radiat Oncol Date: 2022-07-26
Authors: Azadeh Abravan; Gareth Price; Kathryn Banfill; Tom Marchant; Matthew Craddock; Joe Wood; Marianne C Aznar; Alan McWilliam; Marcel van Herk; Corinne Faivre-Finn Journal: Front Oncol Date: 2022-07-19 Impact factor: 5.738
Authors: Andrew Hope; Maikel Verduin; Thomas J Dilling; Ananya Choudhury; Rianne Fijten; Leonard Wee; Hugo Jwl Aerts; Issam El Naqa; Ross Mitchell; Marc Vooijs; Andre Dekker; Dirk de Ruysscher; Alberto Traverso Journal: Cancers (Basel) Date: 2021-05-14 Impact factor: 6.639