Predicting tumour motion during the whole radiotherapy treatment: a systematic approach for thoracic and abdominal lesions based on real time MR

Davide Cusumano, Jennifer Dhont, Luca Boldrini, Giuditta Chiloiro, Stefania Teodoli, Mariangela Massaccesi, Bruno Fionda, Francesco Cellini, Luigi Azario, Jef Vandemeulebroucke, Marco De Spirito, Vincenzo Valentini, Dirk Verellen

Research output: Contribution to journalArticle

Abstract

Introduction: Aim of this study was to investigate the ability of pre-treatment four dimensional computed tomography (4DCT) to capture respiratory-motion observed in thoracic and abdominal lesions during treatment. Treatment motion was acquired using full-treatment cine-MR acquisitions. Results of this analysis were compared to the ability of 30 seconds (s) cine Magnetic Resonance (MR) to estimate the same parameters. Methods: A 4DCT and 30 s cine-MR (ViewRay, USA) were acquired on the simulation day for 7 thoracic and 13 abdominal lesions. Mean amplitude, intra- and inter-fraction amplitude variability, and baseline drift were extracted from the full treatment data acquired by 2D cine-MR, and correlated to the motion on pre-treatment 30 s cine-MR and 4DCT. Using the full treatment data, safety margins on the ITV, necessary to account for all motion variability from 4DCT observed during treatment, were calculated. Mean treatment amplitudes were 2 ± 1 mm and 5 ± 3 mm in the anteroposterior (AP) and craniocaudal (CC) direction, respectively. Differences between mean amplitude during treatment and amplitude on 4DCT or during 30 s cine-MR were not significant, but 30 s cine-MR was more accurate than 4DCT. Intra-fraction amplitude variability was positively correlated with both 30 s cine-MR and 4DCT amplitude. Inter-fraction amplitude variability was minimal. Results: Mean baseline drift over all fractions and patients equalled 1 ± 1 mm in both CC and AP direction, but drifts per fraction up to 16 mm (CC) and 12 mm (AP) were observed. Margins necessary on the ITV ranged from 0 to 8 mm in CC and 0 to 5 mm in AP direction. Neither amplitude on 4DCT nor during 30 s cine MR is correlated to the magnitude of drift or the necessary margins in both directions. Conclusion: Lesions moving with small amplitude show limited amplitude variability throughout treatment, making passive motion management strategies seem adequate. However, other variations such as baseline drifts and shifts still cause significant geometrical uncertainty, favouring real-time monitoring and an active approach for all lesions influenced by respiratory motion.

Original languageEnglish
Pages (from-to)456-462
Number of pages7
JournalRadiotherapy and Oncology
Volume129
Issue number3
DOIs
Publication statusPublished - Dec 1 2018

Fingerprint

Magnetic Resonance Spectroscopy
Radiotherapy
Thorax
Neoplasms
Therapeutics
Four-Dimensional Computed Tomography
Uncertainty
Safety
Direction compound

Keywords

  • Inter-fraction variability
  • Intra-fraction fraction variability
  • Motion prediction
  • MR-guided radiotherapy

ASJC Scopus subject areas

  • Hematology
  • Oncology
  • Radiology Nuclear Medicine and imaging

Cite this

Predicting tumour motion during the whole radiotherapy treatment : a systematic approach for thoracic and abdominal lesions based on real time MR. / Cusumano, Davide; Dhont, Jennifer; Boldrini, Luca; Chiloiro, Giuditta; Teodoli, Stefania; Massaccesi, Mariangela; Fionda, Bruno; Cellini, Francesco; Azario, Luigi; Vandemeulebroucke, Jef; De Spirito, Marco; Valentini, Vincenzo; Verellen, Dirk.

In: Radiotherapy and Oncology, Vol. 129, No. 3, 01.12.2018, p. 456-462.

Research output: Contribution to journalArticle

Cusumano, Davide ; Dhont, Jennifer ; Boldrini, Luca ; Chiloiro, Giuditta ; Teodoli, Stefania ; Massaccesi, Mariangela ; Fionda, Bruno ; Cellini, Francesco ; Azario, Luigi ; Vandemeulebroucke, Jef ; De Spirito, Marco ; Valentini, Vincenzo ; Verellen, Dirk. / Predicting tumour motion during the whole radiotherapy treatment : a systematic approach for thoracic and abdominal lesions based on real time MR. In: Radiotherapy and Oncology. 2018 ; Vol. 129, No. 3. pp. 456-462.
@article{3d4d982d7c494f7e9a014d7d6766061f,
title = "Predicting tumour motion during the whole radiotherapy treatment: a systematic approach for thoracic and abdominal lesions based on real time MR",
abstract = "Introduction: Aim of this study was to investigate the ability of pre-treatment four dimensional computed tomography (4DCT) to capture respiratory-motion observed in thoracic and abdominal lesions during treatment. Treatment motion was acquired using full-treatment cine-MR acquisitions. Results of this analysis were compared to the ability of 30 seconds (s) cine Magnetic Resonance (MR) to estimate the same parameters. Methods: A 4DCT and 30 s cine-MR (ViewRay, USA) were acquired on the simulation day for 7 thoracic and 13 abdominal lesions. Mean amplitude, intra- and inter-fraction amplitude variability, and baseline drift were extracted from the full treatment data acquired by 2D cine-MR, and correlated to the motion on pre-treatment 30 s cine-MR and 4DCT. Using the full treatment data, safety margins on the ITV, necessary to account for all motion variability from 4DCT observed during treatment, were calculated. Mean treatment amplitudes were 2 ± 1 mm and 5 ± 3 mm in the anteroposterior (AP) and craniocaudal (CC) direction, respectively. Differences between mean amplitude during treatment and amplitude on 4DCT or during 30 s cine-MR were not significant, but 30 s cine-MR was more accurate than 4DCT. Intra-fraction amplitude variability was positively correlated with both 30 s cine-MR and 4DCT amplitude. Inter-fraction amplitude variability was minimal. Results: Mean baseline drift over all fractions and patients equalled 1 ± 1 mm in both CC and AP direction, but drifts per fraction up to 16 mm (CC) and 12 mm (AP) were observed. Margins necessary on the ITV ranged from 0 to 8 mm in CC and 0 to 5 mm in AP direction. Neither amplitude on 4DCT nor during 30 s cine MR is correlated to the magnitude of drift or the necessary margins in both directions. Conclusion: Lesions moving with small amplitude show limited amplitude variability throughout treatment, making passive motion management strategies seem adequate. However, other variations such as baseline drifts and shifts still cause significant geometrical uncertainty, favouring real-time monitoring and an active approach for all lesions influenced by respiratory motion.",
keywords = "Inter-fraction variability, Intra-fraction fraction variability, Motion prediction, MR-guided radiotherapy",
author = "Davide Cusumano and Jennifer Dhont and Luca Boldrini and Giuditta Chiloiro and Stefania Teodoli and Mariangela Massaccesi and Bruno Fionda and Francesco Cellini and Luigi Azario and Jef Vandemeulebroucke and {De Spirito}, Marco and Vincenzo Valentini and Dirk Verellen",
year = "2018",
month = "12",
day = "1",
doi = "10.1016/j.radonc.2018.07.025",
language = "English",
volume = "129",
pages = "456--462",
journal = "Radiotherapy and Oncology",
issn = "0167-8140",
publisher = "Elsevier Ireland Ltd",
number = "3",

}

TY - JOUR

T1 - Predicting tumour motion during the whole radiotherapy treatment

T2 - a systematic approach for thoracic and abdominal lesions based on real time MR

AU - Cusumano, Davide

AU - Dhont, Jennifer

AU - Boldrini, Luca

AU - Chiloiro, Giuditta

AU - Teodoli, Stefania

AU - Massaccesi, Mariangela

AU - Fionda, Bruno

AU - Cellini, Francesco

AU - Azario, Luigi

AU - Vandemeulebroucke, Jef

AU - De Spirito, Marco

AU - Valentini, Vincenzo

AU - Verellen, Dirk

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Introduction: Aim of this study was to investigate the ability of pre-treatment four dimensional computed tomography (4DCT) to capture respiratory-motion observed in thoracic and abdominal lesions during treatment. Treatment motion was acquired using full-treatment cine-MR acquisitions. Results of this analysis were compared to the ability of 30 seconds (s) cine Magnetic Resonance (MR) to estimate the same parameters. Methods: A 4DCT and 30 s cine-MR (ViewRay, USA) were acquired on the simulation day for 7 thoracic and 13 abdominal lesions. Mean amplitude, intra- and inter-fraction amplitude variability, and baseline drift were extracted from the full treatment data acquired by 2D cine-MR, and correlated to the motion on pre-treatment 30 s cine-MR and 4DCT. Using the full treatment data, safety margins on the ITV, necessary to account for all motion variability from 4DCT observed during treatment, were calculated. Mean treatment amplitudes were 2 ± 1 mm and 5 ± 3 mm in the anteroposterior (AP) and craniocaudal (CC) direction, respectively. Differences between mean amplitude during treatment and amplitude on 4DCT or during 30 s cine-MR were not significant, but 30 s cine-MR was more accurate than 4DCT. Intra-fraction amplitude variability was positively correlated with both 30 s cine-MR and 4DCT amplitude. Inter-fraction amplitude variability was minimal. Results: Mean baseline drift over all fractions and patients equalled 1 ± 1 mm in both CC and AP direction, but drifts per fraction up to 16 mm (CC) and 12 mm (AP) were observed. Margins necessary on the ITV ranged from 0 to 8 mm in CC and 0 to 5 mm in AP direction. Neither amplitude on 4DCT nor during 30 s cine MR is correlated to the magnitude of drift or the necessary margins in both directions. Conclusion: Lesions moving with small amplitude show limited amplitude variability throughout treatment, making passive motion management strategies seem adequate. However, other variations such as baseline drifts and shifts still cause significant geometrical uncertainty, favouring real-time monitoring and an active approach for all lesions influenced by respiratory motion.

AB - Introduction: Aim of this study was to investigate the ability of pre-treatment four dimensional computed tomography (4DCT) to capture respiratory-motion observed in thoracic and abdominal lesions during treatment. Treatment motion was acquired using full-treatment cine-MR acquisitions. Results of this analysis were compared to the ability of 30 seconds (s) cine Magnetic Resonance (MR) to estimate the same parameters. Methods: A 4DCT and 30 s cine-MR (ViewRay, USA) were acquired on the simulation day for 7 thoracic and 13 abdominal lesions. Mean amplitude, intra- and inter-fraction amplitude variability, and baseline drift were extracted from the full treatment data acquired by 2D cine-MR, and correlated to the motion on pre-treatment 30 s cine-MR and 4DCT. Using the full treatment data, safety margins on the ITV, necessary to account for all motion variability from 4DCT observed during treatment, were calculated. Mean treatment amplitudes were 2 ± 1 mm and 5 ± 3 mm in the anteroposterior (AP) and craniocaudal (CC) direction, respectively. Differences between mean amplitude during treatment and amplitude on 4DCT or during 30 s cine-MR were not significant, but 30 s cine-MR was more accurate than 4DCT. Intra-fraction amplitude variability was positively correlated with both 30 s cine-MR and 4DCT amplitude. Inter-fraction amplitude variability was minimal. Results: Mean baseline drift over all fractions and patients equalled 1 ± 1 mm in both CC and AP direction, but drifts per fraction up to 16 mm (CC) and 12 mm (AP) were observed. Margins necessary on the ITV ranged from 0 to 8 mm in CC and 0 to 5 mm in AP direction. Neither amplitude on 4DCT nor during 30 s cine MR is correlated to the magnitude of drift or the necessary margins in both directions. Conclusion: Lesions moving with small amplitude show limited amplitude variability throughout treatment, making passive motion management strategies seem adequate. However, other variations such as baseline drifts and shifts still cause significant geometrical uncertainty, favouring real-time monitoring and an active approach for all lesions influenced by respiratory motion.

KW - Inter-fraction variability

KW - Intra-fraction fraction variability

KW - Motion prediction

KW - MR-guided radiotherapy

UR - http://www.scopus.com/inward/record.url?scp=85052060815&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85052060815&partnerID=8YFLogxK

U2 - 10.1016/j.radonc.2018.07.025

DO - 10.1016/j.radonc.2018.07.025

M3 - Article

C2 - 30144955

AN - SCOPUS:85052060815

VL - 129

SP - 456

EP - 462

JO - Radiotherapy and Oncology

JF - Radiotherapy and Oncology

SN - 0167-8140

IS - 3

ER -