Journal Pre-proof

Quantifying the interfractionalmotion of esophagus using daily cone beam computed

tomography with oral contrast during radiotherapy for locally advanced non-small cell
lung cancer

Bo Qiu, MD, PhD, Shi Pei Lu, MS, Bin Wang, PhD, Jian Hui Shao, MS, Ying Ting
Zhang, MS, Meng Yun Qiang, MD, Jin Yu Guo, MS, Yin Zhou, PhD, Hai Hang Jiang,
PhD, Cheng Guang Lin, MS, Xiao Yan Huang, PhD, Xiao Wu Deng, PhD, Qi Wen Li,
MD, PhD, Hui Liu, MD, PhD
PII: S1879-8500(20)30163-6
DOI: https://doi.org/10.1016/j.prro.2020.06.006
Reference: PRRO 1255

To appear in: Practical Radiation Oncology

Received Date: 29 December 2019

Revised Date: 15 April 2020
Accepted Date: 10 June 2020

Please cite this article as: Qiu B, Lu SP, Wang B, Shao JH, Zhang YT, Yun Qiang M, Guo JY, Zhou
Y, Jiang HH, Lin CG, Huang XY, Deng XW, Li QW, Liu H, Quantifying the interfractionalmotion of
esophagus using daily cone beam computed tomography with oral contrast during radiotherapy
for locally advanced non-small cell lung cancer, Practical Radiation Oncology (2020), doi: https://
doi.org/10.1016/j.prro.2020.06.006.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier Inc. on behalf of American Society for Radiation Oncology.
Quantifying the interfractionalmotion of esophagus using daily cone beam

computed tomography with oral contrast during radiotherapy for locally advanced

non-small cell lung cancer

Bo Qiu, MD, PhD1,2,3,4*,ShiPei Lu, MS1,2,4*, Bin Wang, PhD1,2,4*, JianHui Shao, MS1,2,4,YingTing Zhang,

MS1,2,4, MengYunQiang, MD1,2,4, JinYu Guo, MS1,2,4, Yin Zhou, PhD5,HaiHang Jiang, PhD5,

ChengGuang Lin, MS1,2,4, XiaoYan Huang, PhD1,2,4, XiaoWu Deng, PhD1,2,4, QiWen Li, MD,

PhD1,2,3,4# , Hui Liu, MD, PhD1,2,3,4

1
State Key Laboratory of Oncology in South China; 2Collaborative Innovation Center for Cancer

Medicine; 3GuangDong Association Study of Thoracic Oncology, 4 Department of Radiation

Oncology, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong 510060, P. R. China;

5
EvidanceMedical Technologies Inc., Suzhou, China.

Corresponding Author:

Hui Liu, M.D., Ph.D.

Department of Radiation Oncology, 651 Dongfengdong Road, Sun Yat-sen University Cancer

Center, Guangzhou, Guangdong, P.R. China. Zip code: 510060

Tel: 862087343031 Fax: 862087343492 E-mail address:liuhuisysucc@sina.com

QiWen Li, M.D., Ph.D.

Department of Radiation Oncology, 651 Dongfengdong Road, Sun Yat-sen University Cancer

Center, Guangzhou, Guangdong, P.R. China. Zip code: 510060

Tel: 862087343031 Fax: 862087343492 E-mail address: liqw@sysucc.org.cn

*
Dr. Bo Qiu, Mr. ShiPei Luand Dr. Bin Wang contributed equally to this work.
#
Dr. QiWen Li was the co-corresponding author.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Sun Yat-sen University Cancer Center.

Written informed consent was obtained from patients for the use of their data in clinical

research.

Availability of data and materials

The datasets are backed up on the Research Data Deposit public platform (RDD,

https://www.researchdata.org.cn, approval number: RDDA2019001184) and are available on

reasonable request.
Conflict of interest

The authors declared no conflict of interest.

Funding
This work was supported by Science and Technology Planning Project of Guangdong Province,
China [Grant Number 2016A020215190].

Acknowledgements

Not applicable.
Quantifying the interfractional motion of the esophagus using daily

cone beam computed tomography with oral contrast during

radiotherapy for locally advanced non-small cell lung cancer

Abstract1

Background and Purpose: To quantify the interfractional motion of the esophagus during

fractionated radiotherapy for locally advanced non-small cell lung cancer (NSCLC).

Materials and methods: We registered simulation four-dimensional computed tomography

(4DCT) and daily cone beam CT (CBCT), and documented the motion of the esophagus centroid at

5mm-interval slices in right-left (RL) and anterior-posterior (AP) directions. Oral barium sulfate

was administrated during CBCT to help localize the esophagus. Thirty-five patients were enrolled.

Thirty-five 4DCT scans, 595 CBCT scans and 25,970 slices were analyzed. The slice-derived motion

values for all patients were presented as 2.5~97.5 percentiles and ranges stratified by segments.

The magnitude of motion for each individual patient was defined as the standard deviation (SD)

of daily motion values stratified by segments. Correlations between the magnitude of motion and

clinical variables were explored.

Results: The 2.5~97.5 percentiles of RL/AP motion were -4.2~7.1/-4.4~5.1, -10.3~6.0/-4.3~3.8,

-8.7~5.5/ -6.4~2.8 and -9.1~4.7/-5.8~3.3mm for cervical, proximal, middle and distal thoracic

esophagus respectively. The interfractional motion was direction- and location-dependent. The

magnitude of RL motion was greater than that of AP motion for the four segments (p<0.05). In RL

direction, the magnitude of motion was greater for middle thoracic esophagus than for cervical

(median SD, 2.7 vs. 2.0 mm, p=0.001) and proximal thoracic esophagus (median SD, 2.7 vs. 2.1

mm, p=0.002). Patients with right lung tumor and bulky lymph nodes tended to display greater RL

1
Abbreviations: anterior-posterior (AP); cone beam computed tomography (CBCT); four-dimensional computed
tomography (4DCT); gastro-esophageal junction (GEJ); maximum intensity projection (MIP); organs at risk (OAR);
planning OAR volumes (PRVs); radiation therapy (RT); standard deviation (SD); right-left (RL)
esophageal motion.

Conclusions: The interfractional motion of the esophagus can be considerable during

radiotherapy in locally advanced NSCLC, especially for middle thoracic esophagus in RL direction.

Strategies to minimize the effect of interfractional esophageal motion on dosimetry should be

considered.

Keywords

Non-small cell lung cancer; radiotherapy; interfractional esophageal motion

Introduction

Radiation therapy (RT) is the backbone of treatment for locally advanced non-small cell lung

cancer (NSCLC) patients who are not surgical candidates [1]. As the utilization of

intensity-modulated radiotherapy (IMRT) improved conformality and limited normal tissue

irradiation, the normal tissue mobility becomes increasingly important[2]. The International

Commission on Radiation Units and Measurements Report (ICRU) 62 has suggested the use of

margins around organs at risk (OAR) to produce planning OAR volumes (PRVs) to account for

geometric uncertainty in the treatment process[3].

The esophagus is a critical organ for dose constraints in RT for lung cancer. The incidence of

severe acute esophagitis (grade 3 or higher) was reported to be about 8.7%-20% in NSCLC

patients treated with IMRT-based concurrent chemoradiotherapy [4, 5]. Previous studies

suggested that severe esophagitis usually leads to poor nutrition and interruption of treatment,

which adversely affects long-term outcome[6, 7]. The use of an esophagus-sparing technique can

reduce the volume of and dose to the esophagus that is irradiated [8-10]. Consistently, the NCCN

guideline version 3. 2019 for NSCLC constraints doses to esophagus more strictly than before and

recommends the use of contralateral esophagus-sparing[11].

Imaging guided RT using bony mark registration reduces the set-up error. Nevertheless, the intra-

and interfractional motion of the esophagus relative to bony structures may still exist and

compromise the dosimetry. The extent of intrafractional motion of the esophagus has been well

documented in previous studies using four-dimensional computed tomography (4DCT)[12-16].

Edith et al. assessed intrafractional esophageal motion in 29 patients with thoracic tumors. They

found margins that would have incorporated all movement in right-left (RL) /anterior-posterior

(AP) directions were 5/5 mm in the proximal esophagus, 7/6 mm in the mid-esophagus, and 9/8

mm in the distal esophagus[12]. Peng J. et al. found the maximum magnitudes of intrafractional

RL/AP motion was 2.4/3.0 mm for the proximal esophagus, 2.7/3.2mm for the middle esophagus

and 10.0/7.0mm for the distal esophagus[13]. However, studies on interfractional esophageal

motion throughout the RT course have been limited, especially in locally advanced NSCLC. Randi
et al. evaluated the interfractional esophageal motion in the RL/AP directions using CT-on-rails

imaging in 8 esophageal cancer patients with a total of 31 CT sets. They found motion was

greatest in the RL direction above the carina. Coverage of 95% of esophageal motion requires

12mm left, 8mm right, 10mm posterior, and 9mm anterior margins. This study provided

important information on the interfractional esophageal motion. However, the sample size did

not permit subset analysis of esophageal mobility in relation to different esophageal location.

What is more, the detailed description of CT acquisition and esophageal delineation was lacking

in their report[17].

The purpose of this study was to quantify the interfractional esophagus motion during the course

of fractionated RT and provided data for motion management strategies for locally advanced

NSCLC. We used enhanced 4DCT for treatment simulation and daily cone beam CT (CBCT) during

treatment delivery to document movement of the esophagus. Specifically, in order to accurately

identify the position of the esophagus on non-enhanced CBCT, oral barium sulfate was

administrated before and during the acquisition of CBCT to clearly define the esophagus lumen.

Materials and Methods

Patient selection

A total of 35 patients with stage IIIA/B NSCLC were enrolled from a prospective clinical trial on

hypofractionated radiotherapy. Radiation was delivered using IMRT to a dose of 51Gy in 17 daily

fractions (NCT02573506). This study was approved by the Ethics Committee of our center.

Written informed consent was obtained from patients for the use of their data in clinical

research.

CT simulation

Patients were immobilized using a vacuum bag in supine position. A simulation 4DCT scan was

obtained with 0.5 cm thickness slices from the Atlas to the second lumbar vertebra level to cover

the whole neck and lung. The CT scan was performed using the Real-time Position Management

system respiratory gating hardware to record the breathing pattern (Varian Medical Systems, Palo
Alto, CA). The CT images were automatically sorted into 10 phases corresponding to a single

respiratory cycle. The maximum intensity projection (MIP) images were reconstructed from the

10 phases of images. Intravenous contrast media administration was mandatory for better

visualization.

Cone beam CT acquiring

A kilovoltage (kV)-based CBCT for daily RT setup was acquired using an Elekta VersaHD system

(Elekta Lts, Crawley, UK) for each patient. Oral barium contrast was administrated before and

during the acquisition of CBCT to help identify the position of the esophagus. The barium sulfate

suspension was diluted to a concentration of 50% wt/vol[18]. In order to delineate the entire

length of lumen, patients were asked to swallow 30 ml of barium before the set-up, and another

30 ml slowly during the acquisition of CBCT in supine position. The CBCT images were acquired

under free breathing with a slice thickness of 5 mm and a maximum scan length of 26.7

centimeters (120 kV, 40 mA, 40 ms, 640frames, 360°data collection) (Figure 1).

Esophagus Contouring

A total of 35 4DCT scans and 595 CBCT scans were obtained for esophagus contouring. The

external esophageal wall was contoured from esophageal orifice to gastro-esophageal junction

(GEJ) under the mediastinal window setting (window width, 400 HU; window level, 40 HU) on the

MIP image of 4DCT. The lumen of the esophagus, as indicated by barium sulfate, was contoured

on each of the daily CBCTs. Esophagus contouring was manually performed by one radiation

oncologist and reviewed by another senior radiation oncologist. Geometric centroids were

automatically placed at the center of the esophagus contour at every slice on the simulation scan

and daily CBCT scan using Zeus TPS V1.0 (Tongdiao, SuZhou, China).

Definition of interfractional esophagus motion

The interfractional motion of the esophagus was defined as the motion between the esophagus

centroid on the simulation CT and that on the daily CBCT scans (Figure 2). All daily CBCT images

were registered to the simulation CT using bony landmarks, thus the motion of the esophagus

centroid represents internal variations relative to patient set-up. The registration was manually
performed by one radiation oncologist and reviewed by another senior radiation oncologist. The

motion data were determined automatically for every slice from the esophageal orifice to the GEJ

(5mm interval) in RL and AP directions using Zeus TPS V1.0 (Tongdiao, SuZhou, China). A negative

value in the AP direction represents patients’ posterior motion. A negative value in the RL

direction represents patients’ right motion. As no fiducials were placed in the esophagus,

cranial-caudal esophageal motion was not addressed due to difficulty identifying superior or

inferior esophageal motion. A total of 25,970 slice-derived motion values were determined and

analyzed.

Data presentation and analysis

The interfractional motion of the esophagus was presented as the motion of four segments[19]:

cervical esophagus (from esophageal orifice to sternal notch), proximal thoracic esophagus (from

sternal notch to carina), middle thoracic esophagus (from carina to left lower pulmonary vein),

and distal thoracic esophagus (from left lower pulmonary vein to GEJ). The slice-derived motion

values for all patients were presented as 2.5~97.5 percentiles and ranges stratified by segments.

The magnitude of motion for each individual patient was defined as the standard deviation of

daily motion values stratified by segments. Wilcoxon signed-rank test was used to compare the

magnitude of motion between RL and AP directions, as well as between the four segments.

Correlations between the motion data and clinical variables, including sex, age, location of

primary tumor, presence of bulky mediastinal lymph node, presence of obstructive pneumonitis,

volume of gross tumor and volume of lungs, were explored. Independent t-sample test was

performed to test for correlations between the categorical variable and motion data. The

statistical analysis was performed using SPSS 19.0. A p value <0.5 was considered statistically

significant.

Results

Patient characteristics

Of 35 enrolled patients, the median age at diagnosis was 62 years. Thirty patients (85.7%) were
male and 5 (14.3%) were female. All patients had histologically confirmed NSCLC, including 10

adenocarcinoma (28.6%) and 25 squamous cell carcinoma (71.4%). Fourteen patients (40.0%)

were staged as IIIA and 21 patients (60.0%) as IIIB. The gross tumor was adjacent to cervical

esophagus in 2 (5.7%) patients, to proximal thoracic esophagus in 29 (82.9%) patients, to middle

thoracic esophagus in 30 (85.7%) patients and to distal thoracic esophagus in 3 (8.6%) patients

(the proximity of location was defined as the minimal distance between the esophagus and gross

tumor < 5mm) (Supplementary Figure 1).

Overall interfractional motion of the esophagus

The overall interfractional motion of four segments was presented in Table 1. The 2.5~97.5

percentiles (range) of RL motion were -4.2~7.1 (-12.5~12.4), -10.3~6.0 (-24.0~11.9), -8.7~5.5

(-27.8~13.1) and -9.1~4.7 (-36.3~10.4)mm for cervical, proximal, middle and distal thoracic

esophagus respectively. The incidence of RL motion ≥3mm was 24.9%, 29.1%, 34.4% and 38.3%;

≥5mm was 9.0%, 12.5%, 15.2% and 14.2%; ≥10mm was 0.3%, 2.8%, 1.9% and 2.0%, for

cervical, proximal, middle and distal thoracic esophagus respectively.

The 2.5~97.5 percentiles (range) of AP motion were -4.4~5.1 (-7.6~9.4), -4.3~3.8 (-9.0~7.4),

-6.4~2.8 (-12.7~7.7) and -5.8~3.3 (-12.4~11.2)mm for cervical, proximal, middle and distal

thoracic esophagus respectively. The incidence of AP motion ≥3mm was 11.2%, 13.4%, 21.0%

and 24.7%; ≥5mm was 3.9%, 1.9%, 7.4% and 8.7%; ≥10mm was 0%, 0%, 0.2% and 0.3, for

cervical, proximal, middle and distal thoracic esophagus respectively.

The interfractional esophageal motion was direction- and location-dependent(Figure 3). The

magnitude of RL motion was greater than that of AP motion for cervical (median SD, 2.0 vs.

1.5mm, p=0.001), proximal (2.1 vs. 1.5mm, p=0.001), middle(2.7 vs. 1.7mm, p=0.000) and distal

thoracic esophagus (2.5 vs. 2.0mm, p=0.017). In RL direction, the magnitude of motion were

greater for middle thoracic esophagus than for cervical (2.7 vs. 2.0 mm, p=0.001) and proximal

thoracic esophagus (2.7 vs. 2.1 mm, p=0.002). In AP direction, the magnitude of motion were

greater for distal thoracic esophagus than for other segments (p<0.05).

Predictor of interfractional RL motion

The RL motion showed great interpatient variability. Figure 4 shows two representing cases

showing large magnitude of motion above and below the carina, respectively. Correlations

between the magnitude of RL motion and clinical variables were explored and listed in table 2.

The mobility of proximal thoracic esophagus correlated significantly with the location of primary

tumor. It displayed greater SD of motion in patients with primary tumor in right lung compared to

patients with primary tumor in left lung (mean SD, 2.7 vs. 1.9mm, p=0.043).

The mobility of middle thoracic esophagus correlated significantly with the presence of bulky

mediastinal lymph nodes. It displayed greater SD of motion in patients with bulky mediastinal

lymph nodes compared to patients without bulky lymph nodes (3.7 vs. 2.4mm, p=0.000).

None of these clinical variables was found to be associated with the mobility of cervical or distal

thoracic esophagus.

Discussion

Esophagus is a critical organ at risk in radiation for NSCLC. Risk factors for radiation esophagitis

have been extensively explored. Most studies found the severity of radiation esophagitis was

significantly associated with dosimetric factors, such as V50 (volume of the esophagus receiving

≥50 Gy), V60 and mean dose [20, 21]. In order to reduce the risk of severe esophagitis, the

radiation dose of the esophagus is strictly constrained, and image-guidance was used for

treatment set-up to reduce the position-related motion. However, the internal esophagus motion

during RT course might still exist and compromise appropriate dosimetry. The interfractional

esophageal motion in locally advanced NSCLC has not been well documented in previous research.

Therefore, the purpose of this study was to quantify the interfractional esophageal motion and

provide data on motion as a basis for the application of motion management strategies.

In fractionated RT, the interfractional movement of an organ relative to the simulation position

will compromise the expected dosimetry and therefore is the parameter of interest. The

magnitude of organ motion is usually presented as SD and range of the movement[22]. In this
study, we used simulation 4DCT as reference and daily CBCT during treatment to document the

motion of the esophagus centroid at 5mm-interval slices from the orifice to GEJ. The 35 4DCT

scans, 595 CBCT scans and 25,970 slice-derived motion values in this report represent the largest

analysis of interfractional esophageal mobility.

The position of the esophagus is not easily identifiable in non-enhanced CBCT, especially for the

middle and distal segments, due to the relatively low spatial resolution of images. Oral barium

sulfate was administrated in our study to help localize the esophagus lumen. Patients were asked

to swallow 30 ml of barium before the set-up, and another 30 ml slowly during the acquisition of

CBCT in supine position, to ensure that the barium sulfate could fully opacify the entire length of

the esophagus. We assumed that the centroid of the lumen equates the centroid of the

esophagus.

From the data of all patients, we found considerable magnitude of interfractional motion

throughout the RT course. As indicated by the 2.5~97.5 percentiles of motion data, margins that

would have incorporated 95% of movement in RL/AP directions were -4.2~7.1/ -4.4~5.1mm for

cervical esophagus, -10.3~6.0/-4.3~3.8mm for proximal thoracic esophagus, -8.7~5.5/-6.4~2.8mm

for middle thoracic esophagus and -9.1~4.7/ -5.8~3.3mm for distal thoracic esophagus. Previous

studies also quantified margins for different segments of the esophagus to account for

respiration-related motion[12, 13]. Compared with their results, our study showed a larger RL

margin in proximal and middle thoracic esophagus, implying that the interfractional motion could

arise of other factors except for respiration, such as esophageal peristalsis and tumor

deformation during the RT course[23]

Theoretically, the distal esophagus might display greater motion compared with other segments

due to less limitation of surrounding structures in the lower mediastinum. Previous studies on

esophageal motion in esophageal cancer coincided with this assumption[12, 13]. However, typical

esophageal motion might be disrupted by the adjacent tumor and anatomic change in

fractionated RT for stage III NSCLC. We compare the SD of the motion data between different
segments to determine whether the motion is segment-dependent. In RL direction, the SD of

motion in middle thoracic esophagus exceeded that for cervical and proximal thoracic esophagus,

and was comparable with that for distal thoracic esophagus. In terms of the inter

2.5~97.5percentile range, the motion for proximal and middle thoracic esophagus was

comparable to that for distal thoracic esophagus. In locally advanced NSCLC, gross tumor was

more frequently adjacent to the proximal and middle thoracic esophagus compared with distal

esophagus (82.9% vs 85.7% vs 8.6%). Therefore, the interfractional mobility of proximal and

middle thoracic esophagus is of special clinical significance, and deserves more attention in RT for

locally advanced NSCLC.

Individual analysis found the magnitude of RL motion varied considerably among patients, the

interpatient variability may be caused by specific patient and tumor characteristics. Exploring the

correlation between clinical variables and esophagus mobility could help individualize the

internal margin for the esophagus. According to our results, patients with bulky lymph nodes

tended to display greater esophageal motion. More considerable anatomic changes in

mediastinal structures were supposed to occur in these patients due to tumor shrinkage during

RT course, and the anatomic changes might be the reason for greater esophageal mobility.

The interfractional motion of the esophagus observed in the current study indicates that the

actual esophagus dose might be misestimated in IMRT plans that assume either limited or no

esophageal mobility. Failing to meet an esophageal constraint does not happen very often in

conventional fractionated radiotherapy. However, with the application of advanced RT technique

such as simultaneous modulated accelerated radiation therapy and new concurrent

chemotherapy/immunotherapy, there might be a significant volume change of gross tumor

during RT. The shrinkage of gross tumor, enlarged mediastinal lymph nodes in particular, seemed

to have greater impact on the motion and the exact location of esophagus adjacent to tumor

than physiological movement itself. For this reason, even though the planning esophageal

constraint is quite perfect, the esophagus could still be irradiated with high dose because the
change of its location during RT (Supplementary Figure 2). Several strategies have been reported

to minimize the effect of interfractional motion [22, 24]; each could have a different dosimetric

impact on tumor dose coverage and esophagus sparing. Therefore, the selection of motion

management strategies should depend on different clinical scenarios.

Results from the current study provided evidence for appropriate internal margins in different

directions and segments. According to our data, daily acquisition of pre-treatment images and

the alignment of the esophagus to the simulation CT provided another method for motion

management for patients with tumor next to esophagus and treated with an escalation dose with

curative intent, in case a treatment plan to avoid the PRV of esophagus might sacrifice the

radiation dose of tumor. For significant motion of esophagus or tumors that cannot be fully

accounted for by daily patient setup, adaptive re-planning should be considered. The MR linear

accelerator integrates a radiotherapy accelerator and a diagnostic quality 1.5T MRI, enabling

soft-tissue visualization during the actual radiation delivery for improved targeting. Better

visualization of the tumor and surrounding tissues as well as onboard adaptive-re-planning make

it possible to adapt the dose to the actual anatomy while optimally sparing esophagus and

normal tissues. We also found that where nodes are not targeted next to the esophageal OAR, it

is probably quite straight forward to meet the esophageal constraints and thus the

recommendation of using an esophagus-sparing technique may not be needed or relevant, since

the risk of severe radiation-induced esophagitis would be quite low in such situation. The

dosimetric impact on esophagus and tumor of different motion management strategies warrant

further investigation.

Our study has some limitations. First, cranial-caudal motion of the esophagus was not recorded

due to difficulty identifying superior or inferior esophageal motion. Second,

contouring/registration variations could have influenced the motion data. In order to minimize

the influences, all contours and registrations in the current study were performed by one

radiation oncologist and reviewed by another.

Conclusions

To our knowledge, this is the first manuscript to report interfractional esophageal motion in RT

for locally advanced NSCLC. Our results showed that the interfractional motion can be

considerable, especially for middle thoracic esophagus in RL direction. Strategies to minimize the

effect of interfractional motion on dosimetry should be considered depend on different clinical

scenarios. Further studies with larger numbers of patients are warranted to confirm these results

and to explore their clinical implications.

References

[1] Bezjak A, Temin S, Franklin G, Giaccone G, Govindan R, Johnson ML, et al. Definitive and Adjuvant
Radiotherapy in Locally Advanced Non-Small-Cell Lung Cancer: American Society of Clinical Oncology
Clinical Practice Guideline Endorsement of the American Society for Radiation Oncology
Evidence-Based Clinical Practice Guideline. J Clin Oncol. 2015;33:2100-5.
[2] Chun SG, Hu C, Choy H, Komaki RU, Timmerman RD, Schild SE, et al. Impact of Intensity-Modulated
Radiation Therapy Technique for Locally Advanced Non-Small-Cell Lung Cancer: A Secondary Analysis
of the NRG Oncology RTOG 0617 Randomized Clinical Trial. J Clin Oncol. 2017;35:56-62.
[3] Stroom JC, Heijmen BJ. Geometrical uncertainties, radiotherapy planning margins, and the ICRU-62
report. Radiother Oncol. 2002;64:75-83.
[4] Wei X, Liu HH, Tucker SL, Liao Z, Hu C, Mohan R, et al. Risk factors for acute esophagitis in
non-small-cell lung cancer patients treated with concurrent chemotherapy and three-dimensional
conformal radiotherapy. Int J Radiat Oncol Biol Phys. 2006;66:100-7.
[5] Kwint M, Uyterlinde W, Nijkamp J, Chen C, de Bois J, Sonke JJ, et al. Acute esophagus toxicity in
lung cancer patients after intensity modulated radiation therapy and concurrent chemotherapy. Int J
Radiat Oncol Biol Phys. 2012;84:e223-8.
[6] Cox JD, Pajak TF, Asbell S, Russell AH, Pederson J, Byhardt RW, et al. Interruptions of high-dose
radiation therapy decrease long-term survival of favorable patients with unresectable non-small cell
carcinoma of the lung: analysis of 1244 cases from 3 Radiation Therapy Oncology Group (RTOG) trials.
Int J Radiat Oncol Biol Phys. 1993;27:493-8.
[7] Ma L, Ye W, Li Q, Wang B, Luo G, Chen Z, et al. Subjective Global Assessment (SGA) Score Could Be
a Predictive Factor for Radiation Pneumonitis in Lung Cancer Patients With Normal Pulmonary
Function Treated by Intensity-Modulated Radiation Therapy and Concurrent Chemotherapy. Clin Lung
Cancer. 2018;19:e211-e7.
[8] Al-Halabi H, Paetzold P, Sharp GC, Olsen C, Willers H. A Contralateral Esophagus-Sparing Technique
to Limit Severe Esophagitis Associated With Concurrent High-Dose Radiation and Chemotherapy in
Patients With Thoracic Malignancies. Int J Radiat Oncol Biol Phys. 2015;92:803-10.
[9] Kao J, Pettit J, Zahid S, Gold KD, Palatt T. Esophagus and Contralateral Lung-Sparing IMRT for Locally
Advanced Lung Cancer in the Community Hospital Setting. Front Oncol. 2015;5:127.
[10] Ma L, Qiu B, Li Q, Chen L, Wang B, Hu Y, et al. An esophagus-sparing technique to limit radiation
esophagitis in locally advanced non-small cell lung cancer treated by simultaneous integrated boost
intensity-modulated radiotherapy and concurrent chemotherapy. Radiat Oncol. 2018;13:130.
[11] NCCN Practice Guideline in Non-Small Cell Lung Cancer. Available at: http://
www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. Accessed: January 20, 2019.
[12] Dieleman EM, Senan S, Vincent A, Lagerwaard FJ, Slotman BJ, van Sornsen de Koste JR.
Four-dimensional computed tomographic analysis of esophageal mobility during normal respiration.
Int J Radiat Oncol Biol Phys. 2007;67:775-80.
[13] Jin P, Hulshof MC, de Jong R, van Hooft JE, Bel A, Alderliesten T. Quantification of
respiration-induced esophageal tumor motion using fiducial markers and four-dimensional computed
tomography. Radiother Oncol. 2016;118:492-7.
[14] Lorchel F, Dumas JL, Noel A, Wolf D, Bosset JF, Aletti P. Esophageal cancer: determination of
internal target volume for conformal radiotherapy. Radiother Oncol. 2006;80:327-32.
[15] Patel AA, Wolfgang JA, Niemierko A, Hong TS, Yock T, Choi NC. Implications of respiratory motion
as measured by four-dimensional computed tomography for radiation treatment planning of
esophageal cancer. Int J Radiat Oncol Biol Phys. 2009;74:290-6.
[16] Sekii S, Ito Y, Harada K, Kitaguchi M, Takahashi K, Inaba K, et al. Intrafraction esophageal motion in
patients with clinical T1N0 esophageal cancer. Reports of practical oncology and radiotherapy : journal
of Greatpoland Cancer Center in Poznan and Polish Society of Radiation Oncology. 2018;23:398-401.
[17] Cohen RJ, Paskalev K, Litwin S, Price RA, Jr., Feigenberg SJ, Konski AA. Esophageal motion during
radiotherapy: quantification and margin implications. Dis Esophagus. 2010;23:473-9.
[18] Gore RM, Levine MS. Textbook of Gastrointestinal Radiology E-Book: Elsevier Health Sciences;
2014.
[19] Rice TW, Ishwaran H, Ferguson MK, Blackstone EH, Goldstraw P. Cancer of the Esophagus and
Esophagogastric Junction: An Eighth Edition Staging Primer. J Thorac Oncol. 2017;12:36-42.
[20] Singh AK, Lockett MA, Bradley JD. Predictors of radiation-induced esophageal toxicity in patients
with non-small-cell lung cancer treated with three-dimensional conformal radiotherapy. Int J Radiat
Oncol Biol Phys. 2003;55:337-41.
[21] Palma DA, Senan S, Oberije C, Belderbos J, de Dios NR, Bradley JD, et al. Predicting esophagitis
after chemoradiation therapy for non-small cell lung cancer: an individual patient data meta-analysis.
Int J Radiat Oncol Biol Phys. 2013;87:690-6.
[22] Langen KM, Jones DT. Organ motion and its management. Int J Radiat Oncol Biol Phys.
2001;50:265-78.
[23] Lischalk JW, Kole TP, Anjum HM, Obayomi-Davies O, Rashid A, Unger K. Four-dimensional
computed tomography prediction of inter- and intrafractional upper gastrointestinal tumor motion
during fractionated stereotactic body radiation therapy. Pract Radiat Oncol. 2016;6:176-82.
[24] Abbas H, Chang B, Chen ZJ. Motion management in gastrointestinal cancers. Journal of
gastrointestinal oncology. 2014;5:223-35.
Figure captions

Figure 1. The Cone beam CT (CBCT) scan with oral barium sulfate at transverse (A) sagittal (B),

and coronal plane (C). The lumen of the esophagus, as indicated by barium, was delineated with

red line.

Figure 2. Slice based analysis of interfractional esophageal motion. (A) The workflow of motion

analysis for each patient. (B) Interfractional motion was defined as the motion between the

centroid of the esophagus on the simulation CT and that on the daily CBCT scans. Green line: the

outline of the esophagus wall on the simulation CT; Yellow line: the outline of the esophagus

lumen on CBCT. The motion was recorded in the right-left (RL) and anterior-posterior (AP)

directions. A total of 35 4DCT scans, 595 CBCT scans and 25,970 slice-derived motion values were

obtained for motion analysis.

Figure 3.Plots showing the magnitude of interfractional motion (standard deviation of daily

motion values) in cervical, proximal thoracic, middle thoracic and distal thoracic esophagus. The

motion was measured in right-left (A) and anterior-posterior directions (B). The dots represents

the standard deviation of daily motion values for each patient; the bars at the middle represents

the mean value of magnitude of each subgroups.

Figure 4. Two representing cases showing large magnitude of motion above and below the carina,

respectively. A,B: a case showing large magnitude of motion in proximal thoracic esophagus; C, D:

a case showing large magnitude of motion in distal thoracic esophagus. Green line: the outline of

the esophagus on the simulation CT; Yellow line: the outline of the esophagus lumen on CBCT.
Table 1. The overall esophageal motion of four segments for all the 35 patients. The slice-derived motion values were presented as 2.5~97.5 percentiles and ranges
stratified by segments. A negative value represents motion to the patient’s right in the right-left (RL) direction and posterior in the anterior-posterior (AP) direction.
N, the number of slice-derived motion values.

Segment N 2.5%~97.5% Range, mm % ≥3mm % ≥5mm % ≥10mm

percentiles, mm
RL AP RL AP RL AP RL AP RL AP
Cervical esophagus 6170 -4.2~7.1 -4.4~5.1 -12.5~12.4 -7.6~9.4 24.9 11.2 9.0 3.9 0.3 0
Proximal thoracic esophagus 5873 -10.3~6.0 -4.3~3.8 -24.0~11.9 -9.0~7.4 29.1 13.4 12.5 1.9 2.8 0
Middle thoracic esophagus 7835 -8.7~5.5 -6.4~2.8 -27.8~13.1 -12.7~7.7 34.4 21.0 15.2 7.4 1.9 0.2
Distal thoracic esophagus 6092 -9.1~4.7 -5.8~3.3 -36.3~10.4 -12.4~11.2 38.3 24.7 14.2 8.7 2.0 0.3
Table 2. Correlation between the magnitude of motion (standard deviation of individual motion values) and clinical variables. Independent t-sample test was
performed to test for correlations between the categorical variable and motion data. A p value <0.05 was considered statistically significant.

Cervical esophagus Proximal thoracic esophagus Middle thoracic esophagus Distal thoracic esophagus
Mean SD P value Mean SD P value Mean SD P value Mean SD P value
Sex 0.423 0.989 0.925 0.928
Male 2.2 2.1 2.7 2.5
Female 2.0 2.1 2.7 2.5
Age 0.666 0.416 0.983 0.252
< 62 yrs 2.0 2.3 2.7 2.3
>= 62 yrs 2.0 2.0 2.7 2.8
Location of primary tumor 0.380 0.043 0.242 0.679
Right lung 2.2 2.7 3.0 2.4
Left lung 2.0 1.9 2.5 2.6
Location of primary tumor 0.230 0.129 0.504 0.997
Central 2.1 2.4 2.8 2.5
Peripheral 1.9 1.8 2.6 2.5
Bulky mediastinal lymph node 0.951 0.241 0.000 0.816
Yes 2.0 2.5 3.7 2.6
No 2.0 2.0 2.4 2.5
Obstructive pneumonitis 0.218 0.088 0.538 0.080
Yes 2.2 2.6 2.6 2.9
No 1.9 1.8 2.8 2.0
Volume of gross tumor 0.360 0.680 0.059 0.195
>=120 cc 2.1 2.2 3.1 2.8
 <120 cc 1.9 2.1 2.3 2.2
Volume of lungs 0.117 0.347 0.939 0.362
>=3200 cc 1.7 1.9 2.7 2.7
< 3200 cc 2.2 2.2 2.7 2.3
Figure 1.

A B C
Figure 2.

A. B.
A.
Figure 3.
S t a n d a r d d e v ia t io n o f m o tio n (m m )

0
2
4
6
8
10
C
e 12
rv
P ic
ro a
x l
im
a
l
th
o
ra
M c
ic
id
P = 0 .0 0 1

d
le
th
o
ra
P = 0 .0 0 2

c
D ic
is
ta
l
th
o
R ig h t - le f t d ir e c t io n

ra
c
ic

B.
S t a n d a r d d e v ia t io n o f m o tio n (m m )
0
2
4
6
8
10
12

C
e
rv
P ic
ro a
x l
im
a
l
th
o
ra
M c
ic
id
d
le
th
o
P = 0 .0 0 1

ra
P = 0 .0 1 4

c
D ic
is
P = 0 .0 0 0

ta
l
th
o
P = 0 .0 1 4

ra
c
ic
A n t e r io r - p o s t e r io r d ir e c t io n
Figure 4

A. B.

C. D.