893 / 1096
S877
ESTRO 36
_______________________________________________________________________________________________
challenged by patient motion during treatment delivery.
Movement during treatment can be induced by pain. To
prevent movement, individualized vacuum cushions are
commonly used immobilization devices in SBRT. This study
evaluates the impact of the use of a vacuum cushion on
intra-fraction movement in patients during SBRT based on
cone beam CT (CBCT) data and the impact of pain on
motion.
Material and Methods
Intra-fraction motion was measured in two groups of
patients treated with SBRT: 25 patients with spinal
metastases using a vacuum cushion (BlueBAG™, Elekta,
Stockholm, Sweden, n= 56 fractions) and 19 patients with
lung lesions treated without a vacuum cushion (n= 68
fractions). For the purpose of this study, the comparison
group was considered to have a fictive thoracic spine
target volume. Intra-fraction motion was assessed by
registering the post-treatment CBCT scan to the planning
CT, based on the volume of interest around the (fictive)
spine metastasis.
Translations and rotations were
determined based on a CBCT bone density match using the
Elekta Medical Systems XVI software.
Absolute values of
displacements in translations and rotations after each
fraction were calculated. Pain at baseline was registered.
Treatment time was similar in both groups. Statistical
significant differences between the two groups regarding
displacements in all directions were tested with Mann-
Whitney. This test was also performed for movement in
patients with and without pain. Mixed models were used
to analyze the differences in movement between two
groups because of multiple measurements (i.e. fractions)
within patients.
Results
Significant differences in intra-fraction movement were
found in 5 out of 6 directions (Table 1) in favor of the
vacuum cushion. Mixed models confirmed a significant
difference in movement in the longitudinal and vertical
direction and pitch axis between patients treated with and
without a vacuum cushion. Intra-fraction movement of ≥
2mm (employed PTV margin) was less frequent in patients
treated with a vacuum cushion (Figure 1). In patients
treated with a vacuum cushion no significant differences
in movement were observed in painful patients versus
patients without pain (Table 1).
Conclusion
The use of a vacuum cushion resulted in a clinically
relevant decrease of intra-fraction movement in multiple
directions during SBRT, which results in a more accurate
dose delivery. In these patients, pain did not influence
movement during radiotherapy.
EP-1621 Intrafraction errors in cranial radiotherapy
with standard VMAT mask: implications for SRS/SRT.
F. Azoury
1
, D. Nasr
1
, C. El Khoury
1
, N. Khater
1
, J.
Barouky
1
, R. Sayah
1
, N. Farah
1
, S. Achkar
1
, E. Nasr
1
1
Hotel Dieu de France Hospital - Saint Joseph University,
Radiation Oncology, Beirut, Lebanon
Purpose or Objective
Frameless SRS/SRT for intracranial tumors enhances
patient comfort. Reproducibility of setup is assured by
systematic CBCT with 6D couch correction before
treatment delivery; however concerns of intrafraction
motion that could limit the use of frameless reinforced
VMAT masks in SRS/SRT still remain. Here we study the
magnitude of errors in rotations and translations when
using standard reinforced VMAT thermoplastic masks.
Material and Methods
We analyzed 100 fractions of patientswho underwent
cranial VMAT onTrueBeam with Perfect Pitch 6DoF couch
(Varian Medical Systems). The patients were immobilized
with reinforced IMRT Masks (CIVCO) and standard neck
support. For patients requiring additional support to the
neck personalized Accuform cushions (CIVCO) were used.
Initial repositioning using CBCT before every fraction and
6DoF corrections was performed. Post-treatment CBCT
repositioning was also performed to assess intra-fractional
motion. Positional errors in all six directions were
compiled in addition to 3D total vector errors (TVE).
Results
Intrafraction translation errors (cm) for the vertical,
cranio-caudal and lateral directions were: 0.01 ± 0.03,
0.02 ± 0.05 and -0.01 ± 0.04 respectively (Mean± SD). 3D
TVE was 0.07 ± 0.04(Mean± SD). Intrafraction rotational
errors for pitch, roll and rotation were 0.13
0
± 0.33
0
, 0.04
0
±0.18
0
and -0.03
0
± 0.35
0
respectively(Mean± SD) (Table.1).
Boxplots presented in Figure.1 show a small variability of
the TVE with a range of errors when we eliminate outliers
of 0.12cm; however 92% of the cases were within 0.1cm
deviation.
Conclusion
The use of personalized single layer masks with custom
made Accuform cushions produces stable positioning for
use on TrueBeam with Perfect Pitch platforms.
Intrafraction motion showed a mean TVE of 0.07 ± 0.04cm.
These results suggest that a PTV margin of 0.1cm for SRS
cases and 0.2cm for SRT cases is justified to mitigate