Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy

Tanwiwat Jaikuna; Eliana vasquez Osorio; David Azria; Jenny Chang-Claude; Maria carmen De santis; Sara Gutiérrez-Enríquez; Marcel Van herk; Peter Hoskin; Maarten Lambrecht; Zoe Lingard; Petra Seibold; Alejandro Seoane; Elena Sperk; R paul Symonds; Christopher j. Talbot; Tiziana Rancati; Tim Rattay; Victoria Reyes; Barry s. Rosenstein; Dirk De ruysscher; Ana Vega; Liv Veldeman; Adam Webb; Catharine m.l. West; Marianne c. Aznar

doi:10.1016/j.breast.2023.103578

Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy

Tanwiwat Jaikuna, Eliana vasquez Osorio, David Azria, Jenny Chang-Claude, Maria carmen De santis, Sara Gutiérrez-Enríquez, Marcel Van herk, Peter Hoskin, Maarten Lambrecht, Zoe Lingard, Petra Seibold, Alejandro Seoane, Elena Sperk, R paul Symonds, Christopher j. Talbot, Tiziana Rancati, Tim Rattay, Victoria Reyes, Barry s. Rosenstein, Dirk De ruysscherAna Vega, Liv Veldeman, Adam Webb, Catharine m.l. West, Marianne c. Aznar

Research output: Contribution to journal › Article › peer-review

Abstract

Background and Purpose: Normal tissue complication probability (NTCP) models can be useful to estimate the risk of fibrosis after breast-conserving surgery (BCS) and radiotherapy (RT) to the breast. However, they are subject to uncertainties. We present the impact of contouring variation on the prediction of fibrosis.
Materials and methods: 280 breast cancer patients treated BCS-RT were included. Nine Clinical Target Volume (CTV) contours were created for each patient: i) CTV_crop (reference), cropped 5mm from the skin and ii) CTV_skin, uncropped and including the skin, iii) segmenting the 95% isodose (Iso95%) and iv) 3 different auto-contouring atlases generating uncropped and cropped contours (Atlas_skin/Atlas_crop). To illustrate the impact of contour variation on NTCP estimates, we applied two equations predicting fibrosis grade ≥ 2 at 5 years, based on Lyman-Kutcher-Burman (LKB) and Relative Seriality (RS) models, respectively, to each contour. Differences were evaluated using repeated-measures ANOVA. For completeness, the association between observed fibrosis events and NTCP estimates was also evaluated using logistic regression.
Results: There were minimal differences between contours when the same contouring approach was followed (cropped and uncropped). CTV_skin and Atlas_skin contours had lower NTCP estimates (-3.92%, IQR 4.00, p<0.05) compared to CTV_crop. No significant difference was observed for Atlas_crop and Iso95% contours compared to CTV_crop. For the whole cohort, NTCP estimates varied between 5.3% and 49.5% (LKB) or 2.2% and 49.6% (RS) depending on the choice of contours. NTCP estimates for individual patients varied by up to a factor of 4. Estimates from “skin” contours showed higher agreement with observed events.
Conclusion: Contour variations can lead to significantly different NTCP estimates for breast fibrosis, highlighting the importance of standardising breast contours before developing and/or applying NTCP models.

Original language	English
Article number	103578
Journal	The Breast
Volume	72
Early online date	11 Sept 2023
DOIs	https://doi.org/10.1016/j.breast.2023.103578
Publication status	Published - 1 Dec 2023

Keywords

Inter-observer variation
Late effects
Fibrosis
Breast cancer
NTCP modelling

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.breast.2023.103578Licence: CC BY

NIHR Manchester Biomedical Research Centre
Bruce, I. (PI), Lord, G. (CoI), Lennon, R. (CoI), Black, G. (CoI), Wedge, D. (CoI), Morris, A. (CoI), Hussell, T. (CoI), Sharrocks, A. (CoI), Stivaros, S. (CoI), Buch, M. (CoI), Gough, J. (CoI), Kostarelos, K. (CoI), Thistlethwaite, F. (CoI), Kadler, K. (CoI), Barton, A. (CoI), Hyrich, K. (CoI), Mcbeth, J. (CoI), O'Neill, T. (CoI), Vestbo, J. (CoI), Simpson, A. (CoI), Singh, S. (CoI), Smith, J. (CoI), Felton, T. (CoI), Murray, C. (CoI), Griffiths, C. (CoI), Cullum, N. (CoI), Rhodes, L. (CoI), Warren, R. (CoI), Paus, R. (CoI), Dumville, J. (CoI), Viros Usandizaga, A. (CoI), Keavney, B. (CoI), Tomaszewski, M. (CoI), Allan, S. (CoI), Body, R. (CoI), Cartwright, E. (CoI), Heagerty, A. (CoI), Kalra, P. (CoI), Miller, C. (CoI), Rutter, M. (CoI), Smith, C. (CoI), Trafford, A. (CoI), Evans, D. (CoI), Crosbie, E. (CoI), Crosbie, P. (CoI), Harvie, M. (CoI), Howell, S. (CoI), Renehan, A. (CoI), Dive, C. (CoI), Blackhall, F. (CoI), Landers, D. (CoI), Krebs, M. (CoI), Cook, N. (CoI), Clarke, R. (CoI), Taylor, S. (CoI), Jorgensen, C. (CoI), Lorigan, P. (CoI), Jayson, G. (CoI), Valle, J. (CoI), Mccabe, M. (CoI), Armstrong, A. (CoI), Freitas, A. (CoI), Illidge, T. (CoI), Choudhury, A. (CoI), Hoskin, P. (CoI), West, C. (CoI), Van Herk, M. (CoI), Faivre-Finn, C. (CoI), Bristow, R. (CoI), Kirkby, K. (CoI), Birtle, A. (CoI), Mackay, R. (CoI), Radford, J. (CoI), Linton, K. (CoI), Higham, C. (CoI), Munro, K. (CoI), Plack, C. (CoI), Arden Armitage, C. (CoI), Bruce, I. (CoI), Moore, D. (CoI), Saunders, G. (CoI), Stone, M. (CoI), Haddock, G. (CoI), Lewis, S. (CoI), Elliott, R. (CoI), Green, J. (CoI), Lovell, K. (CoI), Morrison, A. (CoI), Shaw, J. (CoI), Bucci, S. (CoI), Ainsworth, J. (CoI), Webb, R. (CoI), Newman, W. (CoI), Banka, S. (CoI), Clayton-Smith, J. (CoI), Payne, K. (CoI), Moldovan, R. (CoI), Wynn, R. (CoI) & Jones, S. (CoI)
1/12/22 → 30/11/27
Project: Research

Cite this

Jaikuna, T., Osorio, E. V., Azria, D., Chang-Claude, J., De santis, M. C., Gutiérrez-Enríquez, S., Van herk, M., Hoskin, P., Lambrecht, M., Lingard, Z., Seibold, P., Seoane, A., Sperk, E., Symonds, R. P., Talbot, C. J., Rancati, T., Rattay, T., Reyes, V., Rosenstein, B. S., ... Aznar, M. C. (2023). Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy. The Breast, 72, Article 103578. https://doi.org/10.1016/j.breast.2023.103578

@article{fa9d40ebe4894592b63417fc10e62d6c,

title = "Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy",

abstract = "Background and Purpose: Normal tissue complication probability (NTCP) models can be useful to estimate the risk of fibrosis after breast-conserving surgery (BCS) and radiotherapy (RT) to the breast. However, they are subject to uncertainties. We present the impact of contouring variation on the prediction of fibrosis.Materials and methods: 280 breast cancer patients treated BCS-RT were included. Nine Clinical Target Volume (CTV) contours were created for each patient: i) CTV_crop (reference), cropped 5mm from the skin and ii) CTV_skin, uncropped and including the skin, iii) segmenting the 95% isodose (Iso95%) and iv) 3 different auto-contouring atlases generating uncropped and cropped contours (Atlas_skin/Atlas_crop). To illustrate the impact of contour variation on NTCP estimates, we applied two equations predicting fibrosis grade ≥ 2 at 5 years, based on Lyman-Kutcher-Burman (LKB) and Relative Seriality (RS) models, respectively, to each contour. Differences were evaluated using repeated-measures ANOVA. For completeness, the association between observed fibrosis events and NTCP estimates was also evaluated using logistic regression.Results: There were minimal differences between contours when the same contouring approach was followed (cropped and uncropped). CTV_skin and Atlas_skin contours had lower NTCP estimates (-3.92%, IQR 4.00, p<0.05) compared to CTV_crop. No significant difference was observed for Atlas_crop and Iso95% contours compared to CTV_crop. For the whole cohort, NTCP estimates varied between 5.3% and 49.5% (LKB) or 2.2% and 49.6% (RS) depending on the choice of contours. NTCP estimates for individual patients varied by up to a factor of 4. Estimates from “skin” contours showed higher agreement with observed events.Conclusion: Contour variations can lead to significantly different NTCP estimates for breast fibrosis, highlighting the importance of standardising breast contours before developing and/or applying NTCP models.",

keywords = "Inter-observer variation, Late effects, Fibrosis, Breast cancer, NTCP modelling",

author = "Tanwiwat Jaikuna and Osorio, {Eliana vasquez} and David Azria and Jenny Chang-Claude and {De santis}, {Maria carmen} and Sara Guti{\'e}rrez-Enr{\'i}quez and {Van herk}, Marcel and Peter Hoskin and Maarten Lambrecht and Zoe Lingard and Petra Seibold and Alejandro Seoane and Elena Sperk and Symonds, {R paul} and Talbot, {Christopher j.} and Tiziana Rancati and Tim Rattay and Victoria Reyes and Rosenstein, {Barry s.} and {De ruysscher}, Dirk and Ana Vega and Liv Veldeman and Adam Webb and West, {Catharine m.l.} and Aznar, {Marianne c.}",

year = "2023",

month = dec,

day = "1",

doi = "10.1016/j.breast.2023.103578",

language = "English",

volume = "72",

journal = "The Breast",

issn = "0960-9776",

publisher = "Churchill Livingstone",

}

Jaikuna, T, Osorio, EV, Azria, D, Chang-Claude, J, De santis, MC, Gutiérrez-Enríquez, S, Van herk, M , Hoskin, P, Lambrecht, M, Lingard, Z, Seibold, P, Seoane, A, Sperk, E, Symonds, RP, Talbot, CJ, Rancati, T, Rattay, T, Reyes, V, Rosenstein, BS, De ruysscher, D, Vega, A, Veldeman, L, Webb, A, West, CML & Aznar, MC 2023, 'Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy', The Breast, vol. 72, 103578. https://doi.org/10.1016/j.breast.2023.103578

TY - JOUR

T1 - Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy

AU - Jaikuna, Tanwiwat

AU - Osorio, Eliana vasquez

AU - Azria, David

AU - Chang-Claude, Jenny

AU - De santis, Maria carmen

AU - Gutiérrez-Enríquez, Sara

AU - Van herk, Marcel

AU - Hoskin, Peter

AU - Lambrecht, Maarten

AU - Lingard, Zoe

AU - Seibold, Petra

AU - Seoane, Alejandro

AU - Sperk, Elena

AU - Symonds, R paul

AU - Talbot, Christopher j.

AU - Rancati, Tiziana

AU - Rattay, Tim

AU - Reyes, Victoria

AU - Rosenstein, Barry s.

AU - De ruysscher, Dirk

AU - Vega, Ana

AU - Veldeman, Liv

AU - Webb, Adam

AU - West, Catharine m.l.

AU - Aznar, Marianne c.

PY - 2023/12/1

Y1 - 2023/12/1

N2 - Background and Purpose: Normal tissue complication probability (NTCP) models can be useful to estimate the risk of fibrosis after breast-conserving surgery (BCS) and radiotherapy (RT) to the breast. However, they are subject to uncertainties. We present the impact of contouring variation on the prediction of fibrosis.Materials and methods: 280 breast cancer patients treated BCS-RT were included. Nine Clinical Target Volume (CTV) contours were created for each patient: i) CTV_crop (reference), cropped 5mm from the skin and ii) CTV_skin, uncropped and including the skin, iii) segmenting the 95% isodose (Iso95%) and iv) 3 different auto-contouring atlases generating uncropped and cropped contours (Atlas_skin/Atlas_crop). To illustrate the impact of contour variation on NTCP estimates, we applied two equations predicting fibrosis grade ≥ 2 at 5 years, based on Lyman-Kutcher-Burman (LKB) and Relative Seriality (RS) models, respectively, to each contour. Differences were evaluated using repeated-measures ANOVA. For completeness, the association between observed fibrosis events and NTCP estimates was also evaluated using logistic regression.Results: There were minimal differences between contours when the same contouring approach was followed (cropped and uncropped). CTV_skin and Atlas_skin contours had lower NTCP estimates (-3.92%, IQR 4.00, p<0.05) compared to CTV_crop. No significant difference was observed for Atlas_crop and Iso95% contours compared to CTV_crop. For the whole cohort, NTCP estimates varied between 5.3% and 49.5% (LKB) or 2.2% and 49.6% (RS) depending on the choice of contours. NTCP estimates for individual patients varied by up to a factor of 4. Estimates from “skin” contours showed higher agreement with observed events.Conclusion: Contour variations can lead to significantly different NTCP estimates for breast fibrosis, highlighting the importance of standardising breast contours before developing and/or applying NTCP models.

AB - Background and Purpose: Normal tissue complication probability (NTCP) models can be useful to estimate the risk of fibrosis after breast-conserving surgery (BCS) and radiotherapy (RT) to the breast. However, they are subject to uncertainties. We present the impact of contouring variation on the prediction of fibrosis.Materials and methods: 280 breast cancer patients treated BCS-RT were included. Nine Clinical Target Volume (CTV) contours were created for each patient: i) CTV_crop (reference), cropped 5mm from the skin and ii) CTV_skin, uncropped and including the skin, iii) segmenting the 95% isodose (Iso95%) and iv) 3 different auto-contouring atlases generating uncropped and cropped contours (Atlas_skin/Atlas_crop). To illustrate the impact of contour variation on NTCP estimates, we applied two equations predicting fibrosis grade ≥ 2 at 5 years, based on Lyman-Kutcher-Burman (LKB) and Relative Seriality (RS) models, respectively, to each contour. Differences were evaluated using repeated-measures ANOVA. For completeness, the association between observed fibrosis events and NTCP estimates was also evaluated using logistic regression.Results: There were minimal differences between contours when the same contouring approach was followed (cropped and uncropped). CTV_skin and Atlas_skin contours had lower NTCP estimates (-3.92%, IQR 4.00, p<0.05) compared to CTV_crop. No significant difference was observed for Atlas_crop and Iso95% contours compared to CTV_crop. For the whole cohort, NTCP estimates varied between 5.3% and 49.5% (LKB) or 2.2% and 49.6% (RS) depending on the choice of contours. NTCP estimates for individual patients varied by up to a factor of 4. Estimates from “skin” contours showed higher agreement with observed events.Conclusion: Contour variations can lead to significantly different NTCP estimates for breast fibrosis, highlighting the importance of standardising breast contours before developing and/or applying NTCP models.

KW - Inter-observer variation

KW - Late effects

KW - Fibrosis

KW - Breast cancer

KW - NTCP modelling

U2 - 10.1016/j.breast.2023.103578

DO - 10.1016/j.breast.2023.103578

M3 - Article

SN - 0960-9776

VL - 72

JO - The Breast

JF - The Breast

M1 - 103578

ER -

Contouring variation affects estimates of normal tissue complication probability for breast fibrosis after radiotherapy

Abstract

Keywords

UN SDGs

Access to Document

Fingerprint

Projects

NIHR Manchester Biomedical Research Centre

Cite this