1

Volumetric Modulated Arc Therapy for Primary Lung Cancer: the Benefit and
Limitations of Including Contralateral Esophagus and Left Anterior Descending Coronary
Artery as Substructure Organs at Risk
Kerry Shroyer RT (R)(T)(CT), Ryan Monago RT(R)(MR)(T), Nishele Lenards, PhD, CMD,
RT(R)(T), FAAMD; Ashley Hunzeker, MS, CMD, Matt Tobler, CMD, R.T.(T) FAAMD
Medical Dosimetry Program at the University of Wisconsin-La Crosse, WI

Introduction
Volumetric modulated arc therapy (VMAT) is an external beam radiation therapy
technique used for curative non-small cell lung cancers (NSCLC) that allows conformal dose
distribution and critical organ sparing. 1One of the primary advantages of VMAT planning is the
ability to spare healthy organs that would otherwise be at risk from other planning techniques.
Volumetric modulated arc therapy lung planning objectives often include whole-organ
constraints such as contralateral lung or heart, but it is not common practice to include the left
anterior descending coronary artery (LAD) and contralateral esophagus (CE) specifically as dose
limiting structures despite clinical research indicating that exceeding dose to these structures can
lead to severe patient complications and morbidity.
The coronary arteries including the LAD originate in the base of the heart.3 Radiation
dose to the base of the heart is associated with non-cancer deaths for patients with NSCLC. 2
Fifteen Gy to 10% or more of the LAD has been associated with an increase in major adverse
cardiac events.4 When considering substructures of the heart, the LAD has the highest
association with adverse outcomes independent from other factors such as cardiac disease
history. Research by Atkins et al4 suggests that the current use of mean dose as the primary
constraint of the heart is limited in value because it does not account for dose distribution across
specific areas of the heart. McWilliam et al4 suggests that contouring and sparing substructures in
the base of the heart rather than the whole heart for dose constraints could lead to better patient
outcomes.
The CE is the esophageal wall region contralateral to the disease and is another structure
not commonly used as an organ at risk (OAR). Al-Halabi et al5 found it is possible to limit total
esophagus cross-section dose for patients with NSCLC receiving intensity modulated radiation
therapy (IMRT) treatments. By reducing the median volume of the CE receiving 45.0 Gy (V 45)
and 55.0 Gy (V55) to 2.5 cubic centimeters (cc) and 0.5 cc respectively, patients were spared of
 2

grade 3 esophagitis which can cause hospital admittance and necessitate a feeding tube. 5-7
Reduction in preventable radiation induced complications could reduce interruptions in patient
treatment courses. Based on previous studies, it can be assumed that considering the CE as a
separate, additional structure rather than the serial organ esophagus, can reduce grade 3
esophagitis as a patient complication.
Literature suggests that careful avoidance of the CE and LAD can lead to better clinical
outcomes for patients The problem is that these structures are not routinely included as standard
avoidance structures for VMAT planning for NSCLC. The purpose of this retrospective planning
study was to determine whether prescription target volume goals could be maintained while
limiting dose to the CE and LAD for VMAT lung planning. Researchers tested the hypotheses
that VMAT lung plans will meet prescription planning tumor volume (PTV) goals while
reducing dose to the LAD to V15<10% (H1A) and reducing dose to the CE to V55Gy<0.5cc (H2A),
V45Gy<2.5cc (H3A), and D0.03cc<60Gy (H4A) as proposed by previous clinical studies.4,5
Methods and Materials
Patient Selection and Setup
Patients selected for this study were previously treated with VMAT plans for primary
lung cancer of various non-small cell histologies. Prescription dose was at least 59.4Gy, 1.8-2.0
Gy per fraction. The patients included had left-side disease or right-side disease with a PTV
within 2.0 cm lateral to the esophagus. Patients were all simulated headfirst, supine, with arms
raised. All patients were scanned with four-dimensional computed tomography (4DCT), 2mm
CT slice thickness, and planned on the average phase series.
Contours
In addition to the previously contoured structures for traditional lung metrics such as
whole hear and, total lung minus internal target volume (ITV), the LAD and CE were contoured
on each patient. The LAD was contoured as an independent structure from the heart and a 1.0
mm planning organ at risk volume (PRV) structure was created for optimization. The CE was
created following the Kamran et al8 guidelines utilized in their phase 1 nonrandomized clinical
trial. First, the gross tumor volume (GTV) created by the physician was examined in the
contouring window. The contralateral esophagus was drawn with a brush tool by contouring the
half of the esophagus distal to the disease on all axial slices where the GTV was present (Images
1-2). Per Al Halabi et al5 guidelines, the contralateral esophagus should not be within 5.0 mm of
 3

the ITV edge. All the cases utilized in this planning study had 5mm expansions on the ITV to
create the PTV, so the defined CE was then cropped out of the PTV to ensure the 5mm distance
to ITV (Image 3).

Treatment Planning
All plans were created using Varian Eclipse version 16.1, 6 MV, and arc geometry was
kept from the original patient treatment so that any changes to dose metrics could be specifically
linked to the addition of new structure constraints. Plans were calculated using AcurosXB16101
algorithm. All plans were created for the same linear accelerator – Varian TrueBeam with a
0.5cm width multi-leaf collimator (MLC) for the inner 20cm section and a 1.0cm MLC width for
the outer 10cm section of MLC. Clinical constraints were unchanged from the previous patient
plan other than the addition of LAD and CE objectives. Optimization goals were adjusted within
planning to address constraints that were exceeding limitations. The Varian optimization tool
“avoid entry” was used for LAD goals. Evaluation of newly optimized plans assessed whether
the retrospective plan still met OAR constraints compared to the previous plan, whether
prescription goals still met, and whether the new plan met the new LAD and CE goals.
Plan Comparison
The retrospective plans were compared to the patient’s original treatment plans by
evaluating specific metrics. First, PTV coverage was compared based on what percentage of the
volume was receiving 100% of prescription dose. All plans were normalized within 0.3% of each
other. The coverage goal was 100% of prescription dose to 95% of the PTV. The overall hotspot
within the plan was also evaluated using 0.03cc volume with a goal of 0.03cc < 110%. Next,
dose to the LAD and CE was compared regarding the goals previously defined. The heart mean
dose was also evaluated, with an ideal constraint of <2000cGy mean dose.
Statistical Analysis
To determine the significance of change in CE and LAD dose outcomes after re-
optimization with the aforementioned dose constraints, the previous values were compared to the
new reported values using a statistical sign-test. The sign test assigned a p-value and p-value to
each set of analyzed data. A p-value <0.10 is considered significant.
Results
Maintenance of PTV coverage
 4

Of the original plans prior to re-optimization, 9 were meeting the goal of V100%≥95%,
meaning that 100% of the prescription dose encompassed at least 95% of the PTV. One plan,
patient 3, was not meeting this constraint at only 91.5%. After re-optimization with the 3 new CE
and LAD dose constraints, all plans maintained their coverage goals, and patient 3 did not
receive any less prescription dose coverage than the original plan. The ability to meet the LAD
and CE goals while maintaining PTV prescription goals rejected the researcher’s null
hypotheses (H10, H20, H30, H40) .

LAD V15Gy ≤ 10%

Prior to reoptimization, 8 of the 10 plans were failing the LAD V15Gy ≤ 10% goal. The
range of values for this constraint were between 0% and 69% with a mean value of 34% and a
standard deviation of 24%. After re-optimization, all 10 plans met this constraint with a new
range of 0% to 9% with a mean value of 4% and a standard deviation of 4% (Figure 4). The sign
test determined the p-value of this constraint to be 0.002, therefore the null hypothesis was
rejected (H10).
CE V55Gy < 0.5cc
Of the 10 retrospective plans, 4 were failing the CE V55Gy goal of <0.5cc prior to this
study. Of the 6 passing plans, 5 had a value of 0 for this category. The plans ranged from 0.0cc to
1.53cc receiving 55Gy, with a mean value of 0.439cc and a standard deviation of 0.571.After re-
optimization, all 10 retrospective plans were under 0.5cc for this category. The range of V 55Gy
was 0.0cc to 0.49cc with a mean of 0.091cc and standard deviation of 0.157 (Figure 5). The sign
test determined the p-value of this constraint to be 0.013, therefore the null hypothesis was
rejected (H20).
CE V45Gy < 2.5cc
Prior to reoptimization, only 1 plan was failing the new V45Gy < 2.5cc constraint at
6.71cc. The range of V45Gy was between 0.0cc to 6.71cc with a mean value of 1.362cc and
standard deviation of 2.039. After re-optimization, all plans met the V45Gy constraint with a
range of 0.0cc to 1.76cc with a mean value of 0.645cc and standard deviation of 0.703 (Figure
6). The sign test determined the p-value of this constraint to be 0.007, therefore the null
hypothesis was rejected (H30). While 9 of the plans were meeting this constraint before this study
with 4 plans at 0.0cc, the 1 plan that exceeded this metric was reduced from 6.71cc to 1.11 cc.
 5

CE D0.03cc <60Gy
Prior to this study, 2 of the 10 plans did not meet the goal of CE D0.03cc < 60Gy. The
range for this constraint was 25.6Gy to 60.8Gy with a mean value of 47.42Gy and a standard
deviation value of 15.22. After re-optimization, all 10 plans met this constraint, with a range
from 24Gy to 59.4Gy, a mean value of 46.7Gy, and a standard deviation value of 12.21 (Figure
7). The sign test determined the p-value of this constraint to be 0.102, indicating this was not a
statistically significant change, however the null hypothesis was rejected because all plans were
able to meet this constraint without compromising PTV coverage (H40).
Discussion
Significant changes to existing OAR constraints as a result of re-optimization
In addition to the new CE and LAD goals, the original OAR goals were re-evaluated to
determine any changes that could be linked directly to the addition of the new LAD and CE
constraints. Heart mean, global max dose, total lung minus ITV mean dose, total lung minus ITV
V20Gy, and total lung minus ITV V5Gy were all compared to determine significant changes. The
initial range of heart mean dose was between 0.74Gy to 19.9Gy. After re-optimization, the heart
mean dose ranged from 2.4Gy to 19.9Gy. Six plans had a reduction in heart mean dose, while 4
plans had an increase. The sign test determined this was not statistically significant with a p-
value of 0.382.
The global max dose goal of V0.03cc receiving <110% was initially met in 8 of the 10
plans and met in 10 of the 10 plans after re-optimization. The p-value for the change in global
dose max was 0.5, indicating it was not statistically significant. It can be assumed based on these
results that adding the CE and LAD constraints did not significantly change the global max dose
to the plan. The total lung minus ITV mean dose saw an insignificant change with a sign test p-
value = 0.103. The total lung minus ITV V20Gy also saw statistically insignificant change with a
sign test p-value of 0.264. The total lung minus ITV V5Gy volume, however, had significant
change with a p-value of 0.006. In 9 of the 10 plans, there was an increase in V5Gy to the total
lung minus ITV.
Summary of Dosimetric Findings
The inclusion of three additional constraints to the existing plans for re-optimization
resulted in a reduction of dose to the LAD and CE for all ten plans while maintaining PTV
coverage, indicating that these constraints are attainable planning goals. There were statistically
 6

significant changes to the LAD V15Gy, CE V55Gy, and CE V45Gy after adding these constraints
into optimization. Changes to other OAR dose were statistically insignificant other than the ITV
V5Gy, which can be attributed to re-directing the low dose of the plan to accommodate these new
treatment goals. The significance of certain OAR constraints such as the low dose lung bath from
VMAT planning may differ based on physician preference, and priority of clinical objectives
must be discussed when planning.
Interpretation
Previous research indicates that the LAD and CE are valuable organs at risk when
assessing potential patient clinical outcomes.2-7 Studies on limiting the V15Gy to the LAD have
found that this treatment goal could lead to better patient outcomes and reduce non-cancer
cardiac morbidity,4 while studies surrounding dose to the CE have found that by reducing the
V45Gy and V55Gy dose, there may be a significant decrease in grade III esophagitis. By
implementing these proposed dose constraints VMAT NSCLC treatments, there is potential for
less cardiac complication and fewer treatment breaks as a result of esophagitis. 5,6
Implementation
Current protocols for VMAT treatment for NSCLC include dose constraints for the
esophagus and heart, but do not indicate dose limitation specifically to the CE or LAD. RTOG
1306, for instance, lists whole esophagus max and mean values alone for esophagus constraints.8
Adding the CE and LAD as a dose limiting structure to existing protocols may aid in better
clinical outcomes.
Limitations
This study was limited to 10 retrospective cases, with tumor volumes within 2cm lateral
to the esophagus, therefore, results cannot be determined for targets outside of that distance. Data
was collected at one clinical facility, using one treatment planning system for one specific
machine. The plans were re-optimized and researchers did not make changes to beam geometry,
therefore further study with variable treatment factors may yield new results.

Conclusion
The CE and LAD are OARs that are not commonly used in VMAT planning for NSCLC
despite research indicating dose should be limited to these structures. Limiting dose to these
structures could improve patient tolerance to treatment and improve patient outcomes. The
 7

purpose of this study was to determine whether prescription target volume goals could be
maintained while limiting dose to the CE and LAD for VMAT lung planning. The addition of
new constraints for the CE and LAD were shown to reduce dose to these structures while not
reducing coverage to the PTV.
The limitations of this study include a small plan population as well as a single treatment
facility performing the re-optimization of the plans. Researching these outcomes on a wider
population of NSCLC VMAT treatments at varying distance from the esophagus and heart could
provide insight into whether these constraints can consistently be met. Further research may
focus on varying success depending on tumor size and distance.
 8

References
1. Chang, J. Intensity-modulated radiotherapy, not 3D conformal, is the preferred technique for
treating locally advanced lung cancer. Semin Radiat Oncol. 2015;25(2)110-116.
https://doi.org/10.1016/j.semradonc.2014.11.004
2. Stam B, Peulen H, Guckenberger M, et al. Dose to heart substructures is associated with non-
cancer death after SBRT in stage I-II NSCLC patients. Radiother Oncol. 2017;123(3):370-
375. http://doi.org/10.1016/j.radonc.2017.04.017
3. McWilliam A, Kennedy J, Hodgson C, Vasquez Osorio E, Faivre-Finn C, van Herk M.
Radiation dose to heart base linked with poorer survival in lung cancer patients. Eur J
Cancer. 2017;85:106-113. http://doi.org/10.1016/j.ejca.2017.07.053
4. Atkins KM, Chaunzwa TL, Lamba N, et al. Association of left anterior descending coronary
artery radiation dose with major adverse cardiac events and mortality in patients with non–
small cell lung cancer. JAMA Oncol. 2021;7(2):206–219.
http://doi.org/10.1001/jamaoncol.2020.6332
5. 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(4):803-810. http://doi.org/10.1016/j.ijrobp.2015.03.018
6. Kamran SC, Yeap BY, Ulysse CA, et al. Assessment of a contralateral esophagus–sparing
technique in locally advanced lung cancer treated with high-dose chemoradiation: a phase 1
nonrandomized clinical trial. JAMA Oncol. 2021;7(6):910–914.
http://doi.org/10.1001/jamaoncol.2021.0281
7. Ma L, Qiu B, Li Q, 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(1):130. https://doi.org/10.1186/s13014-018-1073-3
8. Govindan, Ramaswamy MD. NRG Oncology. A Randomized Phase II Study of
Individualized Combined Modality Therapy for Stage III Non-Small Cell Lung Cancer
(NSCLC). RTOG 1306. Updated July 25, 2017. Accessed September 21, 2023.
https://classic.clinicaltrials.gov/ProvidedDocs/96/NCT01822496/Prot_SAP_000.pdf
 9

Figures

Figure 1. This is an example of the CE created by contouring one half of the esophagus on the
side distal to the treatment volume and limited to slices containing GTV.

Figure 2. This is another example of the CE contour created using the contouring guidelines by
Al Halabi et al.5
 10

Figure 3. This is a third example of the CE contour, showing that the CE structure is cropped
outside of the PTV, thus maintaining a 5mm distance from the ITV.

LAD Volume Receiving 15Gy

VMAT Lung_10
VMAT Lung_9
VMAT Lung_8
VMAT Lung_7
VMAT Lung Plan

VMAT Lung_6
VMAT Lung_5
VMAT Lung_4
VMAT Lung_3
VMAT Lung_2
VMAT Lung_1
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0%
Percent of Contour
Replan Value receiving
Original Plan 15Gy
Value

Figure 4. This chart shows the change in LAD V15Gy dose from the original plans to the re-
optimized plans.
 11

Contralateral Esophagus Volume Receiving 55Gy

1.8
1.6
1.4
1.2
Volume in cubic centimeters (cc)

1
0.8
0.6
0.4
0.2
0
VMAT VMAT VMAT VMAT VMAT VMAT VMAT VMAT VMAT VMAT
Lung_1 Lung_2 Lung_3 Lung_4 Lung_5 Lung_6 Lung_7 Lung_8 Lung_9 Lung_10

VMAT Lung Plan

Original Plan Value Replan Value

Figure 5. This chart shows the change in CE V55Gy dose from the original plans to the re-
optimized plans.

Contralateral Esophagus Volume Receiving 45Gy

8

7
Volume in cubic centimeters (cc)

0
VMAT VMAT VMAT VMAT VMAT VMAT VMAT VMAT VMAT VMAT
Lung_1 Lung_2 Lung_3 Lung_4 Lung_5 Lung_6 Lung_7 Lung_8 Lung_9 Lung_10

Axis Title

Original Plan Value Replan Value

Figure 6. This chart shows the change in CE V45Gy dose from the original plans to the re-
optimized plans.
 12

Contralateral Esophagus .03cc DMAX Dose Comparison

7000

6000

5000

4000
Dose in cGy

3000 Original Plan Value

Replan Value
2000

1000

0
_1 _2 _3 _4 _5 _6 _7 _8 _9 _1
0
ng ng ng ng ng ng ng ng ng ng
Lu Lu Lu Lu Lu Lu Lu Lu Lu Lu
AT AT AT AT AT AT AT AT AT AT
VM VM VM VM VM VM VM VM V M
VM
VMAT Lung Plan

Figure 7. This chart shows the change in CE 0.03cc Dmax dose from the original plans to the re-
optimized plans.
 13

Tables