Dose-mass inverse optimization for minimally moving thoracic lesions

Ivaylo B Mihaylov, E. G. Moros

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

In the past decade, several different radiotherapy treatment plan evaluation and optimization schemes have been proposed as viable approaches, aiming for dose escalation or an increase of healthy tissue sparing. In particular, it has been argued that dose-mass plan evaluation and treatment plan optimization might be viable alternatives to the standard of care, which is realized through dose-volume evaluation and optimization. The purpose of this investigation is to apply dose-mass optimization to a cohort of lung cancer patients and compare the achievable healthy tissue sparing to that one achievable through dose-volume optimization. Fourteen non-small cell lung cancer (NSCLC) patient plans were studied retrospectively. The range of tumor motion was less than 0.5 cm and motion management in the treatment planning process was not considered. For each case, dose-volume (DV)-based and dose-mass (DM)-based optimization was performed. Nine-field step-and-shoot IMRT was used, with all of the optimization parameters kept the same between DV and DM optimizations. Commonly used dosimetric indices (DIs) such as dose to 1% the spinal cord volume, dose to 50% of the esophageal volume, and doses to 20 and 30% of healthy lung volumes were used for cross-comparison. Similarly, mass-based indices (MIs), such as doses to 20 and 30% of healthy lung masses, 1% of spinal cord mass, and 33% of heart mass, were also tallied. Statistical equivalence tests were performed to quantify the findings for the entire patient cohort. Both DV and DM plans for each case were normalized such that 95% of the planning target volume received the prescribed dose. DM optimization resulted in more organs at risk (OAR) sparing than DV optimization. The average sparing of cord, heart, and esophagus was 23, 4, and 6%, respectively. For the majority of the DIs, DM optimization resulted in lower lung doses. On average, the doses to 20 and 30% of healthy lung were lower by approximately 3 and 4%, whereas lung volumes receiving 2000 and 3000 cGy were lower by 3 and 2%, respectively. The behavior of MIs was very similar. The statistical analyses of the results again indicated better healthy anatomical structure sparing with DM optimization. The presented findings indicate that dose-mass-based optimization results in statistically significant OAR sparing as compared to dose-volume-based optimization for NSCLC. However, the sparing is case-dependent and it is not observed for all tallied dosimetric endpoints.

Original languageEnglish (US)
Pages (from-to)3927-3937
Number of pages11
JournalPhysics in Medicine and Biology
Volume60
Issue number10
DOIs
StatePublished - May 21 2015

Fingerprint

Thorax
Lung
Organs at Risk
Non-Small Cell Lung Carcinoma
Spinal Cord
Standard of Care
Articular Range of Motion
Esophagus
Lung Neoplasms
Radiotherapy
Therapeutics
Neoplasms

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

Dose-mass inverse optimization for minimally moving thoracic lesions. / Mihaylov, Ivaylo B; Moros, E. G.

In: Physics in Medicine and Biology, Vol. 60, No. 10, 21.05.2015, p. 3927-3937.

Research output: Contribution to journalArticle

@article{b921912e21834c5fa332ae80c31d5385,
title = "Dose-mass inverse optimization for minimally moving thoracic lesions",
abstract = "In the past decade, several different radiotherapy treatment plan evaluation and optimization schemes have been proposed as viable approaches, aiming for dose escalation or an increase of healthy tissue sparing. In particular, it has been argued that dose-mass plan evaluation and treatment plan optimization might be viable alternatives to the standard of care, which is realized through dose-volume evaluation and optimization. The purpose of this investigation is to apply dose-mass optimization to a cohort of lung cancer patients and compare the achievable healthy tissue sparing to that one achievable through dose-volume optimization. Fourteen non-small cell lung cancer (NSCLC) patient plans were studied retrospectively. The range of tumor motion was less than 0.5 cm and motion management in the treatment planning process was not considered. For each case, dose-volume (DV)-based and dose-mass (DM)-based optimization was performed. Nine-field step-and-shoot IMRT was used, with all of the optimization parameters kept the same between DV and DM optimizations. Commonly used dosimetric indices (DIs) such as dose to 1{\%} the spinal cord volume, dose to 50{\%} of the esophageal volume, and doses to 20 and 30{\%} of healthy lung volumes were used for cross-comparison. Similarly, mass-based indices (MIs), such as doses to 20 and 30{\%} of healthy lung masses, 1{\%} of spinal cord mass, and 33{\%} of heart mass, were also tallied. Statistical equivalence tests were performed to quantify the findings for the entire patient cohort. Both DV and DM plans for each case were normalized such that 95{\%} of the planning target volume received the prescribed dose. DM optimization resulted in more organs at risk (OAR) sparing than DV optimization. The average sparing of cord, heart, and esophagus was 23, 4, and 6{\%}, respectively. For the majority of the DIs, DM optimization resulted in lower lung doses. On average, the doses to 20 and 30{\%} of healthy lung were lower by approximately 3 and 4{\%}, whereas lung volumes receiving 2000 and 3000 cGy were lower by 3 and 2{\%}, respectively. The behavior of MIs was very similar. The statistical analyses of the results again indicated better healthy anatomical structure sparing with DM optimization. The presented findings indicate that dose-mass-based optimization results in statistically significant OAR sparing as compared to dose-volume-based optimization for NSCLC. However, the sparing is case-dependent and it is not observed for all tallied dosimetric endpoints.",
author = "Mihaylov, {Ivaylo B} and Moros, {E. G.}",
year = "2015",
month = "5",
day = "21",
doi = "10.1088/0031-9155/60/10/3927",
language = "English (US)",
volume = "60",
pages = "3927--3937",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "10",

}

TY - JOUR

T1 - Dose-mass inverse optimization for minimally moving thoracic lesions

AU - Mihaylov, Ivaylo B

AU - Moros, E. G.

PY - 2015/5/21

Y1 - 2015/5/21

N2 - In the past decade, several different radiotherapy treatment plan evaluation and optimization schemes have been proposed as viable approaches, aiming for dose escalation or an increase of healthy tissue sparing. In particular, it has been argued that dose-mass plan evaluation and treatment plan optimization might be viable alternatives to the standard of care, which is realized through dose-volume evaluation and optimization. The purpose of this investigation is to apply dose-mass optimization to a cohort of lung cancer patients and compare the achievable healthy tissue sparing to that one achievable through dose-volume optimization. Fourteen non-small cell lung cancer (NSCLC) patient plans were studied retrospectively. The range of tumor motion was less than 0.5 cm and motion management in the treatment planning process was not considered. For each case, dose-volume (DV)-based and dose-mass (DM)-based optimization was performed. Nine-field step-and-shoot IMRT was used, with all of the optimization parameters kept the same between DV and DM optimizations. Commonly used dosimetric indices (DIs) such as dose to 1% the spinal cord volume, dose to 50% of the esophageal volume, and doses to 20 and 30% of healthy lung volumes were used for cross-comparison. Similarly, mass-based indices (MIs), such as doses to 20 and 30% of healthy lung masses, 1% of spinal cord mass, and 33% of heart mass, were also tallied. Statistical equivalence tests were performed to quantify the findings for the entire patient cohort. Both DV and DM plans for each case were normalized such that 95% of the planning target volume received the prescribed dose. DM optimization resulted in more organs at risk (OAR) sparing than DV optimization. The average sparing of cord, heart, and esophagus was 23, 4, and 6%, respectively. For the majority of the DIs, DM optimization resulted in lower lung doses. On average, the doses to 20 and 30% of healthy lung were lower by approximately 3 and 4%, whereas lung volumes receiving 2000 and 3000 cGy were lower by 3 and 2%, respectively. The behavior of MIs was very similar. The statistical analyses of the results again indicated better healthy anatomical structure sparing with DM optimization. The presented findings indicate that dose-mass-based optimization results in statistically significant OAR sparing as compared to dose-volume-based optimization for NSCLC. However, the sparing is case-dependent and it is not observed for all tallied dosimetric endpoints.

AB - In the past decade, several different radiotherapy treatment plan evaluation and optimization schemes have been proposed as viable approaches, aiming for dose escalation or an increase of healthy tissue sparing. In particular, it has been argued that dose-mass plan evaluation and treatment plan optimization might be viable alternatives to the standard of care, which is realized through dose-volume evaluation and optimization. The purpose of this investigation is to apply dose-mass optimization to a cohort of lung cancer patients and compare the achievable healthy tissue sparing to that one achievable through dose-volume optimization. Fourteen non-small cell lung cancer (NSCLC) patient plans were studied retrospectively. The range of tumor motion was less than 0.5 cm and motion management in the treatment planning process was not considered. For each case, dose-volume (DV)-based and dose-mass (DM)-based optimization was performed. Nine-field step-and-shoot IMRT was used, with all of the optimization parameters kept the same between DV and DM optimizations. Commonly used dosimetric indices (DIs) such as dose to 1% the spinal cord volume, dose to 50% of the esophageal volume, and doses to 20 and 30% of healthy lung volumes were used for cross-comparison. Similarly, mass-based indices (MIs), such as doses to 20 and 30% of healthy lung masses, 1% of spinal cord mass, and 33% of heart mass, were also tallied. Statistical equivalence tests were performed to quantify the findings for the entire patient cohort. Both DV and DM plans for each case were normalized such that 95% of the planning target volume received the prescribed dose. DM optimization resulted in more organs at risk (OAR) sparing than DV optimization. The average sparing of cord, heart, and esophagus was 23, 4, and 6%, respectively. For the majority of the DIs, DM optimization resulted in lower lung doses. On average, the doses to 20 and 30% of healthy lung were lower by approximately 3 and 4%, whereas lung volumes receiving 2000 and 3000 cGy were lower by 3 and 2%, respectively. The behavior of MIs was very similar. The statistical analyses of the results again indicated better healthy anatomical structure sparing with DM optimization. The presented findings indicate that dose-mass-based optimization results in statistically significant OAR sparing as compared to dose-volume-based optimization for NSCLC. However, the sparing is case-dependent and it is not observed for all tallied dosimetric endpoints.

UR - http://www.scopus.com/inward/record.url?scp=84929094193&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929094193&partnerID=8YFLogxK

U2 - 10.1088/0031-9155/60/10/3927

DO - 10.1088/0031-9155/60/10/3927

M3 - Article

C2 - 25909516

AN - SCOPUS:84929094193

VL - 60

SP - 3927

EP - 3937

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 10

ER -