Medical Dosimetry 38 (2013) 366–371
Medical Dosimetry
journal homepage: www.meddos.org
Superior liver sparing by combined coplanar/noncoplanar volumetricmodulated arc therapy for hepatocellular carcinoma: A planning and
feasibility study
Yi-Chun Tsai, M.S.,* Chiao-Ling Tsai, M.D.,* Feng-Ming Hsu, M.D.,* Jian-Kuen Wu, M.S.,*║
Chien-Jang Wu, Ph.D.,║ and Jason Chia-Hsien Cheng, M.D., Ph.D.*†‡§
*Division of Radiation Oncology, Department of Oncology, National Taiwan University Hospital, Taipei, Taiwan; †Graduate Institute of Oncology, National Taiwan University College
of Medicine, Taipei, Taiwan; ‡Graduate Institute of Clinical Medicine, National Taiwan University College of Medicine, Taipei, Taiwan; §Cancer Research Center, National Taiwan
University College of Medicine, Taipei, Taiwan; and ║Institute of Electro-Optical Science and Technology, National Taiwan Normal University, Taipei, Taiwan
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 16 October 2012
Accepted 4 April 2013
Compared with step-and-shoot intensity-modulated radiotherapy (sIMRT) and tomotherapy, volumetricmodulated arc therapy (VMAT) allows additional arc configurations in treatment planning and noncoplanar (NC) delivery. This study was first to compare VMAT planning with sIMRT planning, and the
second to evaluate the toxicity of coplanar (C)/NC-VMAT treatment in patients with hepatocellular
carcinoma (HCC). Fifteen patients with HCC (7 with left-lobe and 8 with right-lobe tumors) were planned
with C-VMAT, C/NC-VMAT, and sIMRT. The median total dose was 49 Gy (range: 40 to 56 Gy), whereas
the median fractional dose was 3.5 Gy (range: 3 to 8 Gy). Different doses/fractionations were converted
to normalized doses of 2 Gy per fraction using an α/β ratio of 2.5. The mean liver dose, volume fraction
receiving more than 10 Gy (V10), 20 Gy (V20), 30 Gy (V30), effective volume (Veff), and equivalent
uniform dose (EUD) were compared. C/NC-VMAT in 6 patients was evaluated for delivery accuracy and
treatment-related toxicity. Compared with sIMRT, both C-VMAT (p ¼ 0.001) and C/NC-VMAT (p ¼ 0.03)
had significantly improved target conformity index. Compared with C-VMAT and sIMRT, C/NC-VMAT for
treating left-lobe tumors provided significantly better liver sparing as evidenced by differences in mean
liver dose (p ¼ 0.03 and p ¼ 0.007), V10 (p ¼ 0.003 and p ¼ 0.009), V20 (p ¼ 0.006 and p ¼ 0.01), V30 (p
¼ 0.02 and p ¼ 0.002), Veff (p ¼ 0.006 and p ¼ 0.001), and EUD (p ¼ 0.04 and p ¼ 0.003), respectively. For
right-lobe tumors, there was no difference in liver sparing between C/NC-VMAT, C-VMAT, and sIMRT. In
all patients, dose to more than 95% of target points met the 3%/3 mm criteria. All 6 patients tolerated
C/NC-VMAT and none of them had treatment-related ≥ grade 2 toxicity. The C/NC-VMAT can be used
clinically for HCC and provides significantly better liver sparing in patients with left-lobe tumors.
& 2013 American Association of Medical Dosimetrists.
Keywords:
Volumetric-modulated arc therapy
Hepatocellular carcinoma
Liver
Noncoplanar
Introduction
Even with the inclusion of radiation therapy (RT) in the multimodality treatment for hepatocellular carcinoma (HCC), the tumor
response and disease outcome remain unsatisfactory.1-3 Insufficient dose of radiation delivered to the target hepatic tumor(s) and
heightened sensitivity of the diseased liver to high-dose radiation
have been the main barriers to achieving satisfactory results.4-6
Reprint requests to: Jason Chia-Hsien Cheng, M.D., Ph.D., Division of Radiation
Oncology, Department of Oncology, National Taiwan University Hospital, No. 7,
Chung-Shan South Road, Taipei 10002, Taiwan.
E-mail: [email protected]
Recent efforts to overcome these barriers include development of
higher-intensity stereotactic body RT (SBRT) and improvement in
the knowledge of radiation-induced liver disease (RILD).7-9 Advances in treatment delivery, including intensity-modulated RT
(IMRT) and respiratory control, make dose escalation possible
and liver sparing achievable.
The clinically adopted technique known as volumetricmodulated arc therapy (VMAT) improves target conformity and
organ sparing by use of rotational IMRT and more control points
for intensity optimization.10-12 Compared with helical tomotherapy
(an earlier arc IMRT technique), VMAT can deliver additional noncoplanar arcs of radiation.13 The liver is situated asymmetrically in
the upper abdomen, and is susceptible to damage from even low
0958-3947/$ – see front matter Copyright Ó 2013 American Association of Medical Dosimetrists
http://dx.doi.org/10.1016/j.meddos.2013.04.003
Y-C. Tsai et al. / Medical Dosimetry 38 (2013) 366–371
doses of radiation. Left lobe of liver is more ventrally and superficially located than right lobe. Noncoplanar VMAT may have the
advantage of taking the shorter path in the liver and further confine
the radiation to a well-defined region of the liver. For the certain
ranges of noncoplanar beam paths, the freedom is limited owing to
the possible mechanical collision of gantry head and couch base.
Therefore, it is technically not possible to have the full-arc noncoplanar VMAT. The target conformity may be compromised with
only the asymmetric partial-arc noncoplanar VMAT design. The
complementary beam design with the coplanar VMAT compensates
this defect. In this study, we used the combined coplanar/noncoplanar VMAT, rather than sole noncoplanar VMAT, to demonstrate
its dosimetric advantage in sparing liver and maintaining target
conformity. The purpose of this study was first to compare the stepand-shoot IMRT (sIMRT), coplanar VMAT (C-VMAT), the combined
coplanar/noncoplanar VMAT (C/NC-VMAT) treatment plans of 15
patients with HCC. In the second part of the study, we used the
VMAT treatment plan that provided comparable or better target
conformity (conformity index [CI]) and liver sparing (proportion of
liver receiving certain doses, effective volume [Veff], and equivalent
uniform dose [EUD]) to treat 6 of the 15 patients. The dose delivery
accuracy and treatment-related toxicity were evaluated in these
patients.
Methods and Materials
Patient characteristics
Fifteen patients with HCC (13 males and 2 females; median age, 61 years
[range: 40 to 82]) who were not candidates for conventional therapies (surgery,
transcatheter arterial chemoembolization, percutaneous ethanol injection therapy,
or radiofrequency ablation) and not eligible for or unresponsive to sorafenib,
underwent RT for localized liver tumors from January 2008 to March 2011, and the
data from these patients were retrospectively analyzed. HCC was Barcelona Clinic
Liver Cancer Stage A in 1 patient, Stage B in 6, and Stage C in 8. Only 1 patient with
Stage A HCC, had a recurrence near the diaphragm after chemoembolization and
declined the recommended surgery owing to his old age (82 years). Twelve patients
were diagnosed serologically with type B chronic viral infection, whereas 2 patients
were with type C. All 15 patients were classified as having Child-Pugh class A
cirrhosis of the liver, and also underwent transcatheter arterial chemoembolization
or radiofrequency ablation before RT or both.
The median gross tumor volume (GTV) and liver volume for 15 patients were
91 mL (range: 5 to 705 mL) and 1219 mL (range: 673 to 1812 mL), respectively. The
median ratio (GTV divided by liver volume) was 0.07 (range: 0.01 to 0.40).
Radiation dose was dependent on the volume of the liver irradiated as determined
by the effective volume method and the estimated risk of liver toxicity.14
Patient setup and immobilization
Each patient was placed in the supine position and immobilized with a
specially designed evacuated vacuum bag. To reduce respiratory movement, an
active breathing coordinator device, body-fix device, and body-frame device were
used in 7 patients, 5 patients, and 1 patient, respectively. The goal of respiratory
control is to reduce the craniocaudal diaphragmatic motion amplitude to be 5 mm
or less on fluoroscopy. Besides the original plan for the actual treatment, 3dimensional computed tomography datasets were used to design sIMRT, C-VMAT,
and C/NC-VMAT treatment plans with the same goals and constraints for each
patient.
Dosimetric plans
GTV was defined as the gross tumor volume, visualized by 3-dimensional
computation of contrast-enhanced computed tomography–defined gross tumor
contours. Clinical target volume was defined as the GTV plus a 0.5-cm margin.
Planning target volume (PTV) was defined as the clinical target volume plus a 0.5cm margin on the medial/lateral/ventral/dorsal sides, as well as a 0.5 to 1.0 cm
margin on the cranial/caudal sides to account for daily setup error and organ
motion due to respiration. The median original prescribed dose was 49 Gy (range:
40 to 56 Gy in 6 to 16 fractions), whereas the median dose per fraction was 3.5 Gy
(range: 3 to 8 Gy). All the prescribed doses normalized to the fraction size of 2 Gy
using an α/β ratio of 2.5 are shown in Table 1. The normalized doses formed the
basis of comparisons between techniques and patients. All 3 plans had comparable
target coverage for each patient (4 95% PTV covered by 4 93% of the prescribed
367
Table 1
Patient characteristics and dose fractionations of 15 patients with hepatocellular
carcinoma normalized to 2-Gy fraction size using the α/β ratio of 2.5
Patient
number
Tumor
lobe
GTV
(mL)
Liver volume
(mL)
Normalized dose
(Gy)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Left
Left
Left
Left
Left
Left
Left
Right
Right
Right
Right
Right
Right
Right
Right
704
266
144
65
221
114
5
91
22
11
75
14
42
128
263
1753
1223
1812
1182
1653
1148
673
1292
1558
850
1054
948
1219
1525
800
104.0
56.0
92.0
70.0
44.4
74.7
56.0
70.0
70.0
65.3
74.7
70.0
65.3
65.3
65.3
dose in 3 patients and by 4 95% in 12 patients). The volume of normal liver was
calculated by subtracting the GTV from the volume of whole liver.
The beam setups of each individual patient in this study were to find the
best achievable designs of the coplanar or noncoplanar beam combination or
both for the 3 techniques. sIMRT plans were designed using the Pinnacle 3
treatment planning system version 9.0 (ADAC Laboratories, Philips Medical
Systems, Milpitas, CA) by the Direct Machine Parameter Optimization algorithm. The minimum segment area was set to 5 cm 2 , and the minimum segment
monitor unit (MU) was 5 MU. The VMAT plans were designed by the SmartArc
module of Pinnacle 3 system for an Elekta Synergy linear accelerator (Elekta
Oncology System Ltd., Crawley, West Sussex, UK), allowing the use of a binned
variable dose rate. Continuous gantry motion, dose-rate variation, and multileaf collimator motion were approximated by optimizing individual beams at 41
gantry angle increments with multileaf collimator leaf positions varying by up
to 4.6 mm for every 11 of gantry rotation. Elekta VMAT delivery was basically by
MU-based servo control. The accelerator used automatic dose-rate selection
that ensures that the maximal possible dose rate was chosen for each
individual segment of the arc. The possible dose rates were 440, 222, 112,
and 57 MU/min. Dose grid resolution was 0.4 cm for the inverse planning.
Pinnacle plans were transferred in Radiotherapy Treatment Plans export
protocols through MOSAIQ version 1.6 (IMPAC Medical Systems, Inc., Sunnyvale, CA) to the linear accelerator. Helical Tomotherapy was not used in this
comparison study, for its limited couch rotation and the small-scale noncoplanar VMAT.
All patients were treated with 10-MV photon beams in supine position. The
beams were directed toward the tumor along paths through the smallest liver
volume to minimize the amount of normal liver exposed to even low doses of
radiation. The quantitative estimation was conducted in the process of selecting the
best noncoplanar beam combinations. The maximum dimensions of PTV were used
to design an open field for the volume estimation of irradiated liver by various
noncoplanar beam combinations, both for sIMRT and for VMAT plans. In general,
the commonly used beams were from right, cranial, and ventral to left, caudal, and
dorsal directions. For sIMRT plans, the median number of beams used was 6 (range:
5 to 8), which included both coplanar and noncoplanar beams. The median number
of segments per beam was 7 (range: 2 to 9). For C-VMAT plans, planning was
performed using 2 clockwise-counterclockwise coplanar partial-arc beams, surrounding the involved lobe of liver, in all patients. For C/NC-VMAT plans, 2
clockwise-counterclockwise coplanar partial arcs and 2 ventral clockwisecounterclockwise noncoplanar arcs surrounding the involved lobe of liver were
used in 12 patients, only ventral clockwise-counterclockwise noncoplanar partial
arcs were used in 2 patients, and 1 coplanar partial-arc plus 4 ventral clockwisecounterclockwise noncoplanar arcs were used in 1 patient.
Doses were prescribed to a peripheral covering isodose covering the PTV.
Assuming dose was normalized to this isodose at 100%, the maximal dose can be
120% and the minimum PTV dose 90%. Any dose 4 110% must be within the PTV.
The normal liver was defined as the normal liver volume minus GTV. In all
patients, it was required that there is at least 700 cc of normal liver, and this
volume received less than 15 Gy. No more than 30% of the normal liver received
more than 27 Gy, and no more than 50% of normal liver received over 24 Gy. For
kidneys, no more than 50% of the combined renal volume received 20 Gy or more.
Maximal permitted dose to spinal cord was 37 Gy. Maximal permitted dose to
small bowel, duodenum, and stomach was 42 Gy for any 3 cc volume. When
planning IMRT or VMAT, the iterations were based on both the target goals and
dose constraints of critical structures, with the liver constraints as the priority.
Therefore, the iterations would continue for the better target conformity if liver
dose-volume data allow.
368
Y-C. Tsai et al. / Medical Dosimetry 38 (2013) 366–371
Conformity index of target and dose to normal liver
were 3% dose difference and 3-mm distance to agreement. Γ ≤ 1 was defined as the
verification passing the criteria and satisfying at least 95% of points.
Several dose-to-target and dose-to-normal tissue indexes were calculated from
dose-volume histograms (DVHs), including Vx, conformity index (CI), effective
volume (Veff), and EUD. Vx was the volume fraction receiving more than a certain
dose (x Gy). CI was the ratio of the prescription volume to the target volume, as
defined in the Radiation Therapy Oncology Group (RTOG) radiosurgery guidelines.15 The CI was defined as follows:
Results
CI ¼
CI of target
V PTV =TV PV
V PTV V TV
¼
,
TV PV =V TV
TV PV 2
C-VMAT, C/NC-VMAT, and sIMRT plans were able to meet the
target conformity goal and the dose-volume constraints for OARs.
The mean CI of PTV was 1.4 ⫾ 0.2 for C-VMAT, 1.5 ⫾ 0.4 for C/NCVMAT, and 1.7 ⫾ 0.4 for sIMRT. The CI was significantly lower for
both C-VMAT (p ¼ 0.001) and C/NC-VMAT (p ¼ 0.03) than for
sIMRT in all patients. Compared with sIMRT, both C-VMAT (1.5 ⫾
0.2 vs 1.9 ⫾ 0.4, p ¼ 0.02) and C/NC-VMAT (1.5 ⫾ 0.4 vs 1.9 ⫾ 0.4,
p ¼ 0.03) achieved superior conformity for right-lobe tumors.
Compared with sIMRT, C-VMAT (1.3 ⫾ 0.1 vs 1.5 ⫾ 0.2, p ¼ 0.02)
but not C/NC-VMAT (1.4 ⫾ 0.4 vs 1.5 ⫾ 0.2, p ¼ 0.64) achieved
superior conformity for left-lobe tumors. The difference in CI for
either lobe was not statistically significant between C-VMAT and
C/NC-VMAT (p 4 0.05). The mean MUs of sIMRT, C-VMAT, and C/NCVMAT were 967 ⫾ 612, 1037 ⫾ 428, and 1007 ⫾ 463, respectively.
where VTV was the treatment volume covered by the prescribed dose, VPTV was the
PTV, and TVPV was the part of the PTV within the VTV.16 The effective volume (Veff)
method from Kutcher and Burman17 was used to estimate the equivalent dose and
volume pairs for uniform partial organ irradiation from the DVHs summarizing the
nonuniform irradiation. The Veff was defined as follows:
ΔV ef f ,i ¼ V i
Di
Dref
1n
V ef f ¼ ∑ ΔV ef f ,i
i
where Di was each dose level, Vi was the volume fraction of the organ receiving the
dose Di, Dref was the maximum dose delivered to the organ, and n (¼ 0.97) was the
volume-effect parameter.14 The concept of EUD was used to determine the uniform
dose for any given nonuniform dose distribution that gives the same biological
effect. The EUD was defined as follows:18,19
"
#n
EUD¼ ∑ V i ðDi Þ1=n
i
,
OARs sparing
or
Differences in dose sparing of the functional liver (liver minus
GTV) were assessed among the 3 techniques (Table 2). Liver
sparing by C/NC-VMAT was significantly more effective in all
parameters including mean liver dose, V10, V20, V30, Veff, and
EUD than that by C-VMAT and sIMRT for left-lobe tumors (p o
0.05) but not right-lobe tumors (p ≥ 0.05). In contrast, both
C-VMAT and sIMRT for either left-lobe or right-lobe tumors
achieved similar liver-sparing results (p ≥ 0.05). The mean DVH
by the normalized dose from 3 techniques (sIMRT, C-VMAT, and
C/NC-VMAT) for PTV and liver of all 15 patients was shown in Fig. 1.
For patients with left-lobe tumors, the correlations of the
tumor size with the liver-sparing advantage between techniques
revealed the significant inverse correlations in mean liver dose of
C/NC-VMAT over C-VMAT (correlation coefficient ¼ −0.79, p ¼
0.03) and in EUD of C/NC-VMAT over C-VMAT (correlation coefficient ¼ −0.80, p ¼ 0.03), as well as insignificant correlations in
the other parameters between techniques (p 4 0.05).
EUD ¼ Dref ðV ef f Þn
Statistical analysis
The descriptive statistics of target conformity and organs at risk (OARs) were
compared between patients receiving sIMRT, C-VMAT, and C/NC-VMAT. A paired
Student t-test was used to evaluate the significance of differences in CI of target,
mean liver dose, V10, V20, V30, Veff, EUD for liver minus GTV, and the absolute
volume or volume fraction receiving specific doses for individual organs. p Value
o 0.05 was defined as statistically significant. SPSS software release 17.00 (SPSS,
Inc., Chicago, IL) was used to calculate the statistics.
Dose verification of VMAT
Doses were verified using the MapCHECK 2 device version 5.02.00.02 (Sun
Nuclear Corporation, Melbourne, FL). The differences between the planned and
measured doses were analyzed by gamma tests. The criteria of gamma evaluation
Table 2
Parameter comparison of liver minus gross tumor volume (Liver−GTV) between step-and-shoot intensity-modulated radiation therapy (sIMRT), coplanar volumetricmodulated arc therapy (C-VMAT), and combined coplanar/noncoplanar VMAT (C/NC-VMAT) for all 15 patients, 7 patients with left-lobe tumors, and 8 patients with rightlobe tumors
Tumor location
(no.)
Liver-GTVparameter
Mean ⫾ standard deviation
sIMRT
All tumors
(n ¼ 15)
Left-lobe tumor
(n ¼ 7)
Right-lobe tumor
(n ¼ 8)
MLD (Gy)
V10 (%)
V20 (%)
V30 (%)
Veff (%)
EUD (Gy)
MLD (Gy)
V10 (%)
V20 (%)
V30 (%)
Veff (%)
EUD (Gy)
MLD (Gy)
V10 (%)
V20 (%)
V30 (%)
Veff (%)
EUD (Gy)
19.7
41.1
30.2
24.7
24.3
19.8
20.5
41.3
30.8
24.1
24.7
20.4
19.0
40.9
29.6
25.2
24.0
19.3
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
C-VMAT
7.5
13.4
9.9
9.9
7.1
7.5
9.4
11.8
11.5
11.7
7.9
9.4
6.0
15.4
9.1
8.7
6.9
6.0
18.9
36.4
27.7
23.1
23.3
19.2
20.0
37.7
28.5
23.5
23.6
20.2
17.9
35.3
27.0
22.8
23.1
18.4
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
8.8
13.0
12.2
11.9
7.9
8.6
10.4
12.3
12.6
13.1
8.1
10.5
7.7
14.2
12.7
11.7
8.3
7.2
p Value
C/NC-VMAT
17.5
33.7
25.3
21.0
21.5
17.8
17. 6
32.0
24.0
19.8
20.6
17.8
17.3
35.2
26.6
22.1
22.2
17.8
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
⫾
7.8
10.9
10.3
10.2
6.7
7.6
9.5
10.7
11.0
11.3
7.1
9.6
6.6
11.6
10.3
9.9
6.7
6.1
MLD ¼ mean liver dose; Vx ¼ volume fraction receiving more than certain dose (x Gy).
C/NC-VMAT vs C-VMAT
C/NC-VMAT vs sIMRT
C-VMAT vs sIMRT
0.02
0.03
0.02
0.02
0.007
0.02
0.03
0.003
0.006
0.02
0.006
0.04
0.31
0.93
0.68
0.45
0.31
0.32
0.002
0.04
0.002
o0.001
0.004
0.004
0.007
0.009
0.01
0.002
0.001
0.003
0.10
0.37
0.11
0.05
0.24
0.17
0.23
0.21
0.10
0.12
0.33
0.42
0.43
0.11
0.26
0.53
0.31
0.74
0.36
0.42
0.26
0.18
0.60
0.49
Y-C. Tsai et al. / Medical Dosimetry 38 (2013) 366–371
Table 3
The treatment-related toxicity in 6 patients treated by volumetric-modulated arc
therapy
Toxicity
Grade 0
Grade I
Grade II
Grade III
Grade IV
Liver
Blood
Coagulation
Skin
Anorexia
Esophagitis
Gastritis
Diarrhea
4
2
5
4
3
5
5
6
2
0
0
2
3
0
1
0
0
1
1
0
0
1
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
For other OARs (lung, stomach, spinal cord, duodenum, bowels,
and kidneys), there were no significant dose differences between
C/NC-VMAT, C-VMMAT, and sIMRT plans (data not shown).
Dose-escalation potentials by C/NC-VMAT than sIMRT
To estimate the dose-escalation potentials by C/NC-VMAT, we
calculated the escalated prescribed dose in the original C/NCVMAT plans but kept the same mean liver dose as that in the
original sIMRT plans for patients with left-lobe tumors. As a result
of the advantageous liver sparing by C/NC-VMAT, the mean
escalated prescribed dose normalized to the fraction size of 2 Gy
369
was 84.5 ⫾ 26.7 Gy compared with the original 71.0 ⫾ 21.2 Gy (p ¼
0.016). The mean dose escalation ratio was 19.4%.
Dose verification and treatment by VMAT
The gamma criteria by 3 measurements for each patient (3%/
3 mm for more than 95% of dose points [mean: 97.7 ⫾ 2.2%; range:
95% to 100%]) were fulfilled in all 6 patients actually treated with
C/NC-VMAT.
Treatment toxicity after C/NC-VMAT
All 6 patients tolerated C/NC-VMAT, and none of them had
treatment-related toxicities greater than grade 2 or RILD within
3 months after completion of treatment, based on Common
Toxicity Criteria version 4.0. The exceptions were 3 patients with
preexisting grade III hematological toxicity due to hypersplenism
before VMAT. Two had low platelet counts and 1 had low
leukocyte counts. All 3 patients had blood counts that were no
worse during and after VMAT. The most common side effects
during VMAT and within 3 months after completion of VMAT were
grade I anorexia (3 patients), grade I dermatitis (2 patients), and
grade I increase in liver enzymes (2 patients), respectively
(Table 3). The average dose delivery time was 382 ⫾ 109 seconds
(range: 243 to 545 seconds).
Fig. 1. The mean dose-volume histogram by the normalized dose from 3 techniques (sIMRT, C-VMAT, and C/NC-VMAT) for (A) planning target volume (PTV) and (B) liver
minus gross tumor volume (Liver−GTV) of all 15 patients.
370
Y-C. Tsai et al. / Medical Dosimetry 38 (2013) 366–371
Fig. 2. Dose distributions, beam illustration, and dose-volume histogram (DVH) of liver minus the gross tumor volume between (A) step-and-shoot intensity-modulated
radiation therapy (sIMRT) (thin solid line on DVH), (B) coplanar volumetric-modulated arc therapy (C-VMAT) (thin dashed line on DVH), (C) combined coplanar/noncoplanar
VMAT (C/NC-VMAT) (thick solid line on DVH), and (D) DVH of liver minus gross tumor volume (Liver−GTV) for a representative patient (no. 3 in Table 1) with left-lobe tumor
of hepatocellular carcinoma.
The dose distributions and DVH of liver minus the GTV for a
representative patient with HCC (no. 3 in Table 1) with a left-lobe
tumor planned by sIMRT, C-VMAT, and C/NC-VMAT techniques are
shown in Fig. 2.
Discussion
In this study, CI of the treated tumor was superior when
delineated by the VMAT technique than by the sIMRT technique
in 15 patients with HCC. More importantly, liver sparing by C/NCVMAT, C-VMAT, and sIMRT was significantly different only for left
hepatic lobe tumors, based on assessment of several parameters
(mean liver dose, V10, V20, V30, Veff, and EUD). Compared with
C-VMAT and sIMRT, C/NC-VMAT significantly improved liver sparing. For patients with right-lobe tumor, both VMAT designs and
sIMRT plans had similar characteristics of dose delivery to the
target and liver. Additionally, 6 patients were actually treated by C/
NC-VMAT with verification of satisfactory dose distribution, and
none of them had excessive hepatic toxicity. To our knowledge,
ours is the first report proposing the use of the VMAT technique,
especially VMAT with the integrated noncoplanar partial-arc
design, for patients with HCC.
RT is one of the modalities used in multidisciplinary
approaches for treating HCC.4,20 The dose to the hepatic tumor
can be suboptimal or if too high may increase the risk of RILD.
SBRT has the advantages of delivering higher dose intensity in
fewer fractions of larger fractional doses and of lowering risk of
RILD by protecting cytokine production from the effects of simultaneous radiation exposure.21 Several parameters, such as volume
fraction of liver receiving certain doses, mean liver dose, Veff, and
EUD, have been shown to be effective in predicting RILD.22-24
Dose-escalation trials also use parameter-driven criteria for dose
selection.7
VMAT has been increasingly used in a variety of disease sites.
The benefits of VMAT are improved target conformity, reduced
MUs (partly by 10-MV photon energy), and shorter delivery time
especially in patients undergoing SBRT.10 Liver malignancies are
now commonly treated by SBRT,7 potentially with advantages by
VMAT. Besides, liver has 2 differently shaped lobes and an
asymmetric position in the upper abdomen. Anatomically, the left
hepatic lobe is more superficially located than right lobe, and is
more exposed to noncoplanar beams from the ventral side. This
advantage is supported by data showing that liver sparing by
C/NC-VMAT is better in patients with left-lobe tumors.
It is of note that the target conformity (CI) of HCC was not as
satisfactory as other disease sites. For most Asian HCC patients
with chronic viral hepatitis and preexisting cirrhosis of liver, it is
important to spare noncancerous liver from even low doses of RT.
Thus, the beam/arc design may be constrained by limited paths
with the smallest volume of liver exposed to radiation. The larger
CI of HCC is likely from this unique situation. Given only the
limited ranges of noncoplanar beam paths, the compromised
freedom to avoid the possible mechanical collision makes the CI
of C/NC-VMAT design acceptably larger than C-VMAT in this study.
Either VMAT or helical tomotherapy (both arc-based
approaches) with more gantry angles than sIMRT can be used for
dose planning. The unique use of noncoplanar arcs by VMAT is
different from tomotherapy (which usually has the highest possible modulation depth) in modulating dose distribution, and makes
possible the delivery of noncoplanar arcs by changing couch
positions. Liver is more susceptible to partial-arc plans because
Y-C. Tsai et al. / Medical Dosimetry 38 (2013) 366–371
of its asymmetric shape, eccentric location, and relationship to the
adjacent duodenum and colon at the hepatic flexure. Therefore,
noncoplanar arcs from the right cranial direction may focus more
of the radiation on the left hepatic lobe, but spare the dorsally and
caudally seated right hepatic lobe and bowels from high doses. The
simple addition of noncoplanar beams into sIMRT might not be as
good in target conformity as arc therapy for the fewer beams by
sIMRT than VMAT. Our evidence shows that C-VMAT or C/NCVMAT, compared with sIMRT, achieves superior target conformity
exclusively with partial arcs and significantly better liver sparing in
patients with left-lobe tumors.
There were a few limitations in this study. Some commercially
available planning systems, including the one used in this study, do
not allow partial-arc plans to be performed with less than 901 of
gantry rotation. Consequently, the selected partial-arc design,
especially one involving the use of integrated noncoplanar beams,
might produce suboptimal results. Besides, the current 2dimensional quality assurance systems have certain directional
limits on accurate dose measurement, which might result in
inconsistency of noncoplanar plans and require the modified
measurement by collapsing the gantry/couch angle to 0 or attaching the array to the gantry. However, such a modified measurement
is not the same dynamic arc delivery as VMAT. Besides, the smaller
dose grid than 4 mm might have finer resolution and potentially
generate different results. Also the available VMAT system cannot
be combined with an active respiratory control device. The inevitable respiratory motion was controlled exclusively by passive
abdominal compression in our patients undergoing VMAT. The
ongoing development of gated VMAT or smaller arc VMAT possibly
integrated with an active breathing coordinator may soon provide
the potential solution. The actual treatment time (including the
time required for MU delivery, the setup procedure, and the image
guidance process) was not compared between techniques. Lastly
the limited number of patients in this study might be associated
with selection bias and confound the comparison.
In conclusion, compared with sIMRT, VMAT provided superior
target conformity in patients treated for HCC. C/NC-VMAT provided
significantly better liver sparing than did C-VMAT and sIMRT in
patients with left-lobe tumors but not right-lobe tumors. Treatment
of HCC with C/NC-VMAT is feasible, accurate, and tolerable.
Acknowledgments
This work was supported by the research grants from National
Science Council, Execute Yuan (NSC 99-2628-B-002-071-MY3 and
99-2314-B-002-111), and National Taiwan University Hospital
grants NTUH 99S1361, 99N1425, 100S1552, and Liver Disease
Prevention & Treatment Research Foundation, Taiwan, ROC.
References
1. Cheng, J.C.H.; Chuang, V.P.; Cheng, S.H.; et al. Local radiotherapy with or
without transcatheter arterial chemoemboliziation for patients with unresectable hepatocellular carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 47:435–42; 2000.
2. Sugahara, S.; Nakayama, H.; Fukuda, K.; et al. Proton-beam therapy for
hepatocellular carcinoma associated with portal vein tumor thrombosis.
Strahlenther. Onkol. 185:782–8; 2009.
371
3. Zhao, J.D.; Liu, J.; Ren, Z.G.; et al. Maintenance of sorafenib following combined
therapy of three-dimensional conformal radiation therapy/intensity-modulated
radiation therapy and transcatheter arterial chemoembolization in patients
with locally advanced hepatocellular carcinoma: a phase I/II study. Radiat.
Oncol. 5:12; 2010.
4. Krishnan, S.; Dawson, L.A.; Seong, J.; et al. Radiotherapy for hepatocellular
carcinoma: an overview. Ann. Surg. Oncol. 15:1015–24; 2008.
5. Hata, M.; Tokuuye, K.; Sugahara, S.; et al. Proton irradiation in a single fraction
for hepatocellular carcinoma patients with uncontrollable ascites. Strahlenther.
Onkol. 183:411–6; 2007.
6. Hata, M.; Tokuuye, K.; Sugahara, S.; et al. Proton beam therapy for hepatocellular carcinoma patients with severe cirrhosis. Strahlenther. Onkol. 182:713–20;
2006.
7. Dawson, L.A.; Tse, R.V.; Hawkins, M.; et al. Phase I study of individualized
stereotactic body radiotherapy for hepatocellular carcinoma and intrahepatic
cholangiocarcinoma. J. Clin. Oncol. 26:657–64; 2008.
8. Cheng, J.C.H.; Wu, J.K.; Huang, C.M.; et al. Radiation-induced liver disease after
three-dimensional conformal radiotherapy for patients with hepatocellular
carcinoma: dosimetric analysis and implication. Int. J. Radiat. Oncol. Biol. Phys.
54:156–62; 2002.
9. Sahgal, A.; Roberge, D.; Schellenberg, D.; et al. The Canadian Association of
radiation oncology scope of practice guidelines for lung, liver and spine
stereotactic body radiotherapy. Clin. Oncol. R. Coll. Radiol. 24:629–39; 2012.
10. Grills, I.S.; McGrath, S.D.; Matuszak, M.M.; et al. Volumetric modulated
arc therapy for delivery of hypofractionated stereotactic lung radiotherapy:
a dosimetric and treatment efficiency analysis. Radiother. Oncol. 95:153–7;
2010.
11. Roa, D.E.; Schiffner, D.C.; Zhang, J.; et al. The use of rapidarc volumetricmodulated arc therapy to deliver stereotactic radiosurgery and stereotactic
body radiotherapy to intracranial and extracranial targets. Med. Dosim.
37:257–64; 2012.
12. Ali, A.N.; Dhabaan, A.H.; Jarrio, C.S.; et al. Dosimetric comparison of volumetric
modulated arc therapy and intensity-modulated radiation therapy for pancreatic malignancies. Med. Dosim. 37:271–5; 2012.
13. Fiveash, J.B.; Clark, G.M.; Popple, R.A.; et al. Feasibility of single-isocenter
volumetric modulated arc radiosurgery for treatment of multiple brain metastases. Int. J. Radiat. Oncol. Biol. Phys. 76:296–302; 2010.
14. Dawson, L.A.; Eccles, C.; Craig, T. Individualized image guided ISO-NTCP based
liver cancer SBRT. Acta. Oncol. 45:856–64; 2006.
15. Shaw, E.; Kline, R.; Gillin, M.; et al. Radiation-therapy oncology group—radiosurgery quality assurance guidelines. Int. J. Radiat. Oncol. Biol. Phys. 27:1231–9;
1993.
16. Han, C.H.; Liu, A.; Schultheiss, T.E.; et al. Dosimetric comparisons of helical
tomotherapy treatment plans and step-and-shoot intensity-modulated radiosurgery treatment plans in intracranl stereotactic radiosurgery. Int. J. Radiat.
Oncol. Biol. Phys. 65:608–16; 2006.
17. Kutcher, G.J.; Burman, C.; Brewster, L.; et al. Histogram reduction method for
calculating complication probabilities for 3-dimensional treatment planning
evaluations. Int. J. Radiat. Oncol. Biol. Phys. 21:137–46; 1991.
18. Niemierko, A.; Goitein, M. Calculation of normal tissue complication probability
and dose-volume histogram reduction schemes for tissues with a critical
element architecture. Radiother. Oncol. 20:166–76; 1991.
19. Delana, A.; Menegotti, L.; Bolner, A.; et al. Impact of residual setup error on
parotid gland dose in intensity-modulated radiation therapy with or without
planning organ-at-risk margin. Strahlenther. Onkol. 185:453–9; 2009.
20. Seong, J.; Koom, W.S.; Han, K.H.; et al. Is local radiotherapy still valuable for
patients with multiple intrahepatic hepatocellular carcinomas? Int. J. Radiat.
Oncol. Biol. Phys. 77:1433–40; 2010.
21. Chou, C.H.; Chen, P.J.; Jeng, Y.M.; et al. Synergistic effect of radiation and
interleukin-6 on hepatitis B virus reactivation in liver through STAT3 signaling
pathway. Int. J. Radiat. Oncol. Biol. Phys. 75:1545–52; 2009.
22. Dawson, L.A.; Normolle, D.; Balter, J.M.; et al. Analysis of radiation-induced liver
disease using the lyman NTCP model. Int. J. Radiat. Oncol. Biol. Phys. 53:810–21;
2002.
23. Cheng, J.C.H.; Wu, J.K.; Lee, P.C.T.; et al. Biologic susceptibility of hepatocellular
carcinoma patients treated with radiotherapy to radiation-induced liver
disease. Int. J. Radiat. Oncol. Biol. Phys. 60:1502–9; 2004.
24. Jiang, G.L.; Liang, S.X.; Zhu, X.D.; et al. Radiation-induced liver disease in threedimensional conformal radiation therapy for primary liver carcinoma: the risk
factors and hepatic radiation tolerance. Int. J. Radiat. Oncol. Biol. Phys.
65:426–34; 2006.