METHODS
Twenty-one patients who were treated with a radiation dose of 27 Gy in three fractions were replanned.
Both VMAT and DCAT plans with single isocenter were obtained for each patient. Plan quality indices,
cumulative monitor unit (MU) values, maximum dose for organs at risk, and mean dose, V19.6Gy, and
V23.1Gy for healthy brain tissue were compared.
RESULTS
The conformity index (p=0.0002), gradient index (p=0.003), maximum dose for brainstem (p=0.016)
and mean dose (p=0.00007), V19.6Gy (p=0.00006), and V23.1Gy (p=0.00006) values for healthy brain
tissue were significantly superior in the VMAT technique, compared to DCAT technique. In addition,
a trend toward significance for achieving lower maximum dose value to the optic nerves and/or pathway
was observed with VMAT (p=0.073). DCAT provided significantly lower MU values (3097.44 vs.
1479.09; p=0.00006).
CONCLUSION
VMAT was able to provide better target conformity and lower risk of brain radionecrosis at least dosimetrically
in multifraction SRS for patients with multiple brain metastases. DCAT may be chosen in
patients with relatively poor performance status or low tolerance to long-lasting radiotherapy sessions.
Keywords: Dosimetric comparison; dynamic conformal arc therapy; multiple brain metastases; stereotactic radiotherapy; volumetric-modulated arc therapy
Radiation necrosis is a dose-limiting toxicity of
brain SRS. The previous studies have shown the relationship
between the occurrence of radionecrosis after
brain SRS and the volume of healthy brain tissue
exposed to both high and low doses of radiation.[
At present, linear accelerator (LINAC)-based SRS is
commonly used worldwide and represents an important
part of radiotherapy applications for brain metastases.
The integration of modern radiotherapy techniques,
such as intensity-modulated radiotherapy (IMRT)
and volumetric-modulated arc therapy (VMAT) into
clinical practice, has led to significant improvements
in LINAC-based SRS planning. In recent years, there
has been an increased interest in using dynamic conformal
arc therapy (DCAT) technique in stereotactic
radiotherapy. DCAT can be used for brain SRS of single
and even multiple brain metastases.[
This study aimed to compare single-isocenter
VMAT and DCAT techniques for multifraction SRS
for multiple brain metastases, in terms of healthy brain
tissue sparing, critical organ doses, quality indices, and
cumulative monitor unit (MU) values. In this context,
there are very few studies in the literature and nearly all of these studies include patients treated with singlefraction
SRS.[
Patient Selection
Patients with 2-10 brain metastases treated with fractionated
brain SRS in Elekta Versa HD linear accelerator
(Elekta, Crawley, UK) at our institution were
included in this study. The prescribed dose for each
target was 27 Gy in three fractions.
Patients with multiple brain metastases who were treated with radiation dose different from 27 Gy in three fractions were excluded from the study. In addition, none of the patients had single brain metastasis, more than 10 brain metastases, or a history of brain surgery.
Immobilization, Contouring, and Prescription
Each patient was immobilized using a thermoplastic
mask in the supine position and scanned with computed
tomography (CT) from the vertex to the base
of skull with a scan thickness of 1.25 mm. No contrast
agent was used. The obtained planning CT images were
appropriately fused with the contrast-enhanced brain
magnetic resonance (MR) images of the patients.
The brain, brainstem, optic nerves/chiasm, lenses,
eyes, and cohleas were contoured as organs at risk
(OARs) in accordance with the European Particle
Therapy Network consensus-based contouring atlas.
[
Treatment Planning and Templates
Both in VMAT and DCAT plans, the isocenter
was placed at the geometric center of the PTVs. Same
gantry and couch angles were used for each plan, and
they were chosen wisely depending on the localization
of lesions by the same medical physicist. Both DCAT
and VMAT plans were based on five non-coplanar arcs.
These partial non-coplanar arcs had an arc length of
120°, regarding gantry rotation. The couch angles were
10°, 45°, 90°, 315°, and 345°; and their corresponding
gantry start angles were 200° clockwise (CW), 320°
counter-clockwise (CCW), 200° CW, 160° CCW, and
40° CW, respectively. Due to the different localization
of the metastases, collimator angles were different for
each plan. All plans were calculated using Monte Carlo
algorithm. The treatment couch, immobilization devices,
and thermoplastic mask were included in the
dose calculation. Dose distributions were calculated
using the 2 mm grid size, and a Monte Carlo dose calculation
uncertainty of 1% per calculation was used.
SSO was used in all DCAT plans to be able to reach the
higher conformity potential of VMAT. Task group 101
dose constraints for three fractions were considered for
dose limitations for OAR.[
Metrics for Plan Comparison
CIRTOG=The prescription isodose volume/Target volume
DHIRTOG=The maximum dose in the target/The prescription
dose
GIPaddick=The volume covered by half of the prescribed
isodose/Target volume Mean dose (Dmean), V19.6Gy and V23.1Gy for non-
PTV brain, maximum dose (Dmax) to the OARs (brainstem,
lenses, cochlea, and optic nerves/chiasm), and
cumulative MU values were extracted from the TPS
for each plan. The values of these plan quality indices
and dose-volume parameters in the VMAT plans were
compared with their equivalents in the DCAT plans.
Statistical Analysis
Radiotherapy plans of the patients who were treated
with multifraction brain SRS were replanned by the
same medical physicist in the Monaco TPS version 5.0
(Elekta, Crawley, UK). In this way, both VMAT and
DCAT plans were obtained for each patient. Flattening filter-free 6 MV (1400 MU/min) photon beams were
used in all plans.
The plan quality indices including conformity index
(CI), gradient index (GI), and dose homogeneity index
(DHI) were calculated as previously proposed by Radiation
Therapy Oncology Group (RTOG) and Paddick
and Lippitz,[
Statistical analysis was performed using the SPSS software
version 22.0 (SPSS Inc., Chicago, IL, USA). The
Shapiro-Wilk test was used to assess data normality.
Data from the DCAT and VMAT plans were compared
using the Wilcoxon signed-rank test and Student's
paired t-test, for non-parametric and parametric
analyses, respectively. P<0.05 was considered as statistically
significant.
Compared with the DCAT technique, CIRTOG
(p=0.0002) and GIPaddick (p=0.003) were significantly superior
in the VMAT technique. However, no significant
difference was found for DHI between two techniques.
In addition, VMAT was significantly superior to DCAT
in terms of exposed radiation dose of healthy brain tissue
including the non-PTV brain V19.6Gy (p=0.00006), V23.1Gy (p=0.00006), and Dmean (p=0.00007) parameters.
On the other hand, significantly lower MU values
were obtained in the DCAT plans (3097.44 vs. 1479.09;
p=0.00006). A visual comparison of dose distributions
in axial, sagittal, and coronal sections of a patient for
each planning technique is shown in Figure
DCAT: Dynamic conformal arc therapy; VMAT: Volumetric-modulated arc therapy.
The maximum doses to the OARs were also compared. No statistically significant difference was found between two planning techniques, regarding Dmax values to the lenses and cochleas. However, a trend toward significance for achieving lower Dmax value to the optic nerves/pathway was observed with VMAT (p=0.073). Importantly, the VMAT technique showed statistically significant benefit for achieving lower Dmax values on brainstem (p=0.016).
In general, our results were similar to the results in
the previous studies comparing VMAT and DCAT in
brain SRS.[
Obtaining the best possible GI and CI values has
been suggested to achieve the optimal stereotactic
radiotherapy planning.[
Inspired by the recommendations in a recent study
of Milano et al.,[
On the other hand, stereotactic radiotherapy is a
time-consuming treatment strategy. In daily practice,
it may be of great significance to complete the radiotherapy
sessions quickly, especially for patients with
relatively poor general condition or low tolerance to
long-lasting radiotherapy sessions. The shortening
of the treatment time may decrease the undesirable
movements, increase the patient's comfort, relieve patient's
distress, and thus reduce treatment-related uncertainties
in this group of patients.
However, it is clear that many factors such as patient/
tumor characteristics, plan quality indices, and
the capabilities of the radiotherapy device should be
considered to obtain optimal brain SRS plan. While
also recognizing all of this, in some selected patients
with low tolerance for long-lasting radiotherapy sessions, choosing DCAT over VMAT in brain SRS may
help clinicians in daily practice, since DCAT could
provide similar target coverage while reducing the
treatment time by half. In this regard, this study supports
the previous studies reporting that DCAT can
reduce treatment time in brain SRS, when compared
to VMAT.[
Limitations and Strength of the Study
The strength of our study is being the first study
comparing VMAT and DCAT plans in patients with
multiple brain metastases who were treated with single-
isocenter three-fraction SRS.
There are some limitations to this study. We compared
a limited number of plan quality indices and dose-volume
parameters, because the optimal parameters are
not well-known and/or well-defined. Different stratifications
could be applied and different parameters other
than those we have compared could be also assessed.
In addition, a comparison considering the sphericity of the target lesions and their distance from each
other and critical organs may provide further important
data. Due to the nature of the study, it could not
be measured whether the dosimetric superiority of
VMAT over DCAT was reflected or not in a clinical
setting, in terms of toxicity.
Peer-review: Externally peer-reviewed.
Conflict of Interest: All authors declared no conflict of interest.
Ethics Committee Approval: The study was approved by the Human Research Ethics Committee of Istinye University (No: 21?70, Date: 23/08/2021).
Financial Support: This study has received no financial support.
Authorship contributions: Concept - G.T., N.B., H.S., B.T., M.Ö., M.F.; Design - G.T., N.B., M.F.; Supervision - G.T., M.F.; Funding - G.T., N.B., H.S., B.T., M.Ö., M.F.; Materials - G.T., N.B., H.S., B.T., M.Ö., M.F.; Data collection and/or processing - G.T., N.B., H.S., B.T., M.Ö.; Data analysis and/ or interpretation - G.T., N.B., H.S.; Literature search - G.T., B.T., M.Ö.; Writing - G.T., N.B.; Critical review - G.T., M.F.
