METHODS
Fifteen patients with multiple brain metastases who had a tumor diameter of <3.5 cm were included in
this study. IMAT, CK, and HT plans were separately created for each patient. The dose prescription was
defined as 18 Gy in the single fraction.
RESULTS
The D40% of hippocampal (in Gy) averaged 1.63, 1.69, and 0.52 for IMAT, CyberKnife, and Tomotherapy,
respectively. The median hippocampal Dmax (in Gy) averaged 2.81, 4.63, and 1.98, respectively. Some
plans were statically different in terms of critical organ doses, but the results were clinically acceptable.
The mean values of V12 (cc) were found to be 12.6, 38.23, and 37.46 for IMAT, CyberKnife, and Tomotherapy,
respectively, when evaluating the doses taken by healthy brain tissue.
CONCLUSION
Brain radiotherapy is a treatment modality for primary and metastatic lesions. However, after radiotherapy
(even with SRS) damage especially in the hippocampus may cause cognitive impairment and a
decrease in patients" quality of life. Therefore, when the hippocampus is outlined as organs of risk, it can
be protected without compromising PTV coverage. We saw this result in all of three treatment platforms
used in this study.
Keywords: CyberKnife; helical tomotherapy; IMAT; multiple brain lesions; stereotactic radiosurgery
Surgery, whole-brain radiation therapy (WBRT),
hippocampal avoidance WBRT, stereotactic radiosurgery
(SRS), chemotherapy, and various combinations
of these treatment techniques are used in the treatment
of brain lesions.[
Recently, more attention has been paid to the quality
of life of the patient folowing cancer treatment and
efforts have been made to reduce the negative effects of
treatment. Irradiation of the brain, especially the hippocampus
region, can lead to more cognitive deficits,
and as a result the quality of life of patients is significantly
affected. These cognitive functions include basic
functions such as thought, memory, attention, association,
and imagination. Any impairment in cognitive
functions in oncological patients can significantly affect
quality of life after completion of treatment and
may lead to deepening of low mood and the emergence
of depressive episodes.[
Nowadays, SRS treatments are performed by
means of Intensity-Modulated Arc Therapy (IMAT),
CK, or HT methods. These techniques can provide a
high level of local control with lower normal tissue
exposure than conventional radiotherapy and provide
sharp dose reductions at target volume limits. IMAT
is a radiation technique that uses multiple densitymodulated
arcs to give the target volume a highly
compatible dose of radiation. Conformal doses can be
obtained by combining treatment field size aperture,
variation of gantry rotation speed, and dose rate. Noncoplanar
radiotherapy uses a series of radiotherapy
beams, but does not share the same geometric plane.
[
Therefore, in this retrospective study, we performed
hippocampus-protected SRC plans on different treatment
platforms using a single isocenter in patients with
2?5 brain metastases. We evaluated the plan parameters,
brain, and other critical organ doses in these treatment
platforms and tried to determine their superiority
over each other. This study was found ethically appropriate
in Istanbul University Institute of Oncology with
the file number 2017/1355 on October 27, 2017.
DICOM sets of 15 patients with multiple brain lesions
were obtained from the archives of our institute.
Treatment volumes and critical organs were drawn by
a radiation oncologist on CT images of patients, and
hippocampal volumes were added by a radiologist.
There was a median of two metastases per plan, with a
mean single tumor volume of 4.05 cc (range, 1.63?7.6
cc) and total tumor volume of 10.5 cc per plan (range,
4.4?24.1 cc). The mean tumor diameter was 2.03 cm
(range, 1.14-3.5 cm). Detailed patient characteristics
are given in Table
Target Volume Definitions
All patients were treated with CK from 2013 to 2018.
A total of 45 plans were evaluated with IMAT and HT,
which were planned separately for each patient. PTV
was created with 0.2 cm margin on GTV. Image sets
for treatment planning were made on Philips Big Bore
4DCT (Philips Healthcare, Cleveland, OH) using 1
mm slice thickness.
Treatment Plans for IMAT, CK, and HT
6 MV photon beams were used for all treatment methods.
The same PTV and OAR volumes have been created
for all the plans; thus, in all the plans, the same
tumor volumes have been irradiated. As an example
of the plan for all treatment models, the axial sections
of the same patient's IMAT, CK, and HT plans are
given in Figure
IMAT: Intensity-modulated Arc Therapy; CK: CyberKnife; HT: Helical radiotherapy.
For each plan, a total of 18 Gy dose were given to
planning target volume (PTV) in 1 fraction using a
dose of 18 Gy per fraction. Plans were made so that at
least 95% of the PTV volume was treated with a dose
of 18 Gy and at least 100% of the GTV was received a
treatment dose of 18 Gy. For each treatment modality,
initial plans were created with the goal of tumor coverage
without hippocampal dose concerns. Later, plans
were re-optimized according to the maximum point
dose Dmax and dose to 40% of the hippocampi D%40. If
the D40 of either hippocampus was greater than 4.50
Gy or maximum hippocampal point dose (Dmax) was greater than or equal to 6.60 Gy, replanning was performed.
These doses constrain were selected based on
RTOG 0933.[
One of the limitations of our study was the nature
of brain tumors that can form different shapes and sizes
in various parts of the brain. The location of the PTV
and distance from the hippocampus varied among patients.
The closest distance of the tumor to the hippocampus
was 0 cm, and the farthest distance was 8.19
cm. Since hippocampus and PTV intersect in 2 of the
study patients, hippocampus protection did not occur
in these patients.
Treatment Plans for IMAT
Treatment Plans for CK
Treatment Plans for HT
Treatment Plan Parameters for PTV
D2%, D50%, and D98% were analyzed in evaluation
of PTV volumes as described in ICRU 83 guidelines.
The homogeneity index (HI) is calculated as follows:
HI=(D2%? D98%)/D50%. For Conformity Index (CI),
CI=Vri/TV formula was used (Where Vri is the volume
of reference isodose and TV is the treatment volume
covered by reference isodose line). The treatment plan
criteria for PTV we found are given in Table
Brain Dose Parameters
OAR Dose Parameters
Statistical Analysis
Single isocentric IMAT plans were planned on Varian
Eclipse TPS (Varian Medical Systems, Version 15.1,
Palo Alto, CA, USA) using four non-coplanar arcs (1
full, 3 half arc). The dose rate was selected at 600 MU/
min. In IMAT optimization, the algorithm "Photon
Optimizer" was used and in dose calculation Anisotropic
Analytical Algorithm was used and grid size was
selected to 1 mm. Multiple shell volumes have been
created to ensure strict coverage of the 9 Gy volume
around PTV. A single isocenter was placed at the center
of the mass of all targets.
CK plans were prepared using Multiplan version 4.0 (Accuray
Inc., Sunnyvale, CA, USA) treatment planning system.
The plans were prepared using two fixed collimators
depending on PTV size. 6D-skull selected as tracing
method. The dose rate was selected at 600 MU/min.
HT plans were performed in the planning system of
the HDA (Accuracy Inc., Sunnyvale, CA, USA). Plans
were performed using pitch=0.143 and modulation
factor=2. The field width was prepared by selecting
1048 and 2512 depending on the tumor size. The dose
calculations for the HT plans were done using the convolution/
superposition algorithm.
The minimum dose in PTV is Dmin, the maximum
dose in PTV is Dmax, and the average dose for PTV is
Dmean. The dose of any percentage of organ volume is
indicated by Dn%. Vn is volume of brain receiving at
least n Gy of radiation dose.
In a study by Minniti et al.,[
In this study, although the plans focus on preserving
the hippocampus and healthy brain tissue, other critical
organ doses are also important. Therefore, in this
study, the doses of all organs in the brain were calculated
in detail and are given in Table
In the SPSS, the first step was a normalization test to
analyze, whether the data were normally distributed.
For normally distributed parameters, a one-way analysis
of variance was used to find significance. Bonferroni
test was applied for double comparison when
"p" value was smaller than 0.05 as a result of this test.
When the distribution was not normal, Kruskal?Wallis
was used to find significance, then a "Mann?Whitney
U"-test was used to find the significance between
the subjects. The limit of significance was set at 0.5 for
p-value. IBM SPSS version 24.0 (SPSS Inc., IL, USA)
was applied for statistical comparison.
Evaluation of Brain Dose Parameters
For each plan, the dose values and statistical results of
the Brain-PTV volume are given in Table
Evaluation of OAR Doses Parameters
For each plan, dose values and statistical results of hippocampus,
optic nerve, chiasm, brain stem, eye, lens,
cochlea, pituitary, and spinal cord are given in Table 5.
When the maximum point dose of the right hippocampus
was evaluated, there was no significant difference between IMAT and HT, but a significant difference
was found between CK and other techniques. There
was no significant difference in other criteria for hippocampus.
Maximum doses were evaluated in the optic
nerve, chiasma, and brain stem. Although there was
no significant difference between the techniques, the
lowest values were obtained with HT. In eye and lens
doses, the lowest values were obtained with CK.
When the maximum and mean doses for the pituitary and left cochlea were evaluated, the lowest doses were obtained with HT. There was no significant difference between the techniques in the right cochlea (p>0.05). For the maximum point dose of the spinal cord, while there was no significant difference between IMAT and CK a significant difference was found between CK and other techniques. Better results were obtained by HT.
Chang et al.[
Nguyen et al.[
The above studies have shown that WBRT is not suitable
for all patients. Together with all these data, these
results suggest that neurocognitive function can be
better preserved by hippocampal protective plans.[
According to Gondi et al.,[
Furthermore, the risk of radiation damage must be
taken into account when treating multiple brain metastases
with SRS. Minniti et al.[
In a study by the University of Cincinnati and Case
Western Reserve, they determined that when the V12
Gy exceeds 10 cc, the risk of necrosis increases by 50%.
In the USCF study, the risk of necrosis was determined
to be 15% when V12 Gy was between 7 and 35 cc.[
On the other hand, in a study by Zhang et al.[
In our study, mean CI resulted 1.04, 1.38, and 1.34
for IMAT, CK, and HT plans, respectively; the differences between IMAT, CK, and HT were statistically
significant. Differences in CI between IMAT, CK, and
HT are attributable to beam collimation systems and
the different dose delivering. Because, while dose is delivered
by a collimated fan beam along an helical pattern
in HT treatments, the irradiation is performed using
non-coplanar beams coming from different angles
in CK.[
Both IMAT and CK reached a high heterogeneity
(mean HI 1.21, 1.15 and for IMAT and CK, respectively),
while a more homogenous dose distribution (mean
HI 1.08) was observed for HT.
In the comparison of CK and HT performed on
19 patients with single brain metastases by Greto et
al.,[
In our study, in addition to doses of healthy brain
tissue and hippocampus other critical organ doses
were also evaluated. As far as we know, there has been
no study, in which all critical organs (brainstem, optic
nerve, chiasm, eye, lens, hippocampus, cochlea, pituitary,
and spinal cord) were evaluated in dosimetric
studies for patients with multiple brain metastases and
treated with SRS. Table 5 shows that all OAR parameters
in the three treatment techniques meet the criteria
for a safe treatment. Critical organ doses provided the
desired criteria in all treatment techniques, while the
lowest doses were generally obtained with HT. Cozzi et
al.[
When using SRS for multiple brain metastases, hippocampus doses may vary according to patients and techniques used. If the hippocampus is identified as the risk organ, preservation of this structure can be achieved in most cases without damaging the target coverage area. Performing the hippocampus preservation procedure during brain RT can significantly reduce or even prevent cognitive complications. This should be thought of in all patients who would have multiple brain metastases and be treated with SRS. Given the amount of dose that healty brain tissue receives, the volume of the brain that receives a low or medium dose is significant. When the doses taken by the healty brain tissue were evaluated in this study, the best protection was obtained from the IMAT plans. Tumor location, size, number, and volume should not be ignored because they are important parameters in determining the treatment technique. As a result, in our study, acceptable results were found for target and critical organ doses in the planning made with three different devices.
Peer-review: Externally peer-reviewed.
Conflict of Interest: All authors declared no conflict of interest.
Ethics Committee Approval: The study was approved by the Istanbul University Institute of Oncology Clinical Research Ethics Committee (no: 2017/1355, date: 27/10/2017).
Financial Support: None declared.
Authorship contributions: Concept - M.O., G.K.; Design - M.O., G.K.; Supervision - M.O., G.K.; Funding - G.K., E.O.G., N.D.S., Ş.K., M.B.; Materials - N.D.S., Ş.K.; Data collection and/or processing - G.K., E.O.G., N.D.S., Ş.K., M.B.; Data analysis and/or interpretation - M.O., G.K.; Literature search - M.O., G.K.; Writing - M.O., G.K.; Critical review - M.O., G.K., N.D.S.
