216 / 1096
S203
ESTRO 36
_______________________________________________________________________________________________
homogeneity.
Conclusion
The TORUS algorithm is able to automatically generate
trajectories having improved plan quality and delivery
time over standard IMRT and VMAT treatments. TORUS
offers an exciting and promising avenue forward toward
increasing our dynamic capabilities in radiation delivery.
OC-0377 Limited interfractional variabi lity of
respiration-induced tumor motion in esophageal
cancer RT
P. Jin
1
, M.C.C.M. Hulshof
1
, N. Van Wieringen
1
, A. Bel
1
, T.
Alderliesten
1
1
Academic Medical Center, Radiation Oncology,
Amsterdam, The Netherlands
Purpose or Objective
Respiration-induced tumor motion is one of the major
sources of intrafractional uncertainties in esophageal
cancer RT. However, the variability thereof during the
treatment course is unclear. In this study, we investigated
the interfractional variability of respiration-induced
esophageal tumor motion using fiducial markers and 4D-
CBCT.
Material and Methods
We included 24 patients with in total 65 markers
implanted in/around the primary esophageal tumor. Per
patient, a 3D planning CT (pCT) and 7–28 (median: 8) 3D-
CBCTs were acquired. Using the fluoroscopy projection
images of the 3D-CBCTs, 10-breathing-phase 4D-CBCTs
were retrospectively reconstructed. First, for each 4D-
CBCT, the 10 phases were rigidly registered to the pCT
based on the vertebra. Next, each marker in each phase
was registered to its corresponding marker in the pCT to
calculate the peak-to-peak amplitude of the respiration-
induced marker motion and the marker motion trajectory.
The mean and standard deviation (SD) of the peak-to-peak
amplitudes over the treatment course were compared
between the left-right (LR), cranial-caudal (CC), and
anterior-posterior (AP) directions; and between the
proximal, middle, distal esophagus, and proximal
stomach. Further, the SDs of the peak-to-peak amplitudes
and marker positions at the inhalation and exhalation
were calculated to assess the interfractional variability of
amplitude and trajectory shape. The correlation between
the mean peak-to-peak amplitude and these SDs was also
assessed.
Results
Overall, the mean and SD of the peak-to-peak amplitudes
were significantly larger in the CC than in the LR/AP
directions (median of mean[SD] in LR/CC/AP (mm):
2.0[0.6]/6.4[0.9]/2.4[0.7];
p
<0.05, Friedman with
Wilcoxon signed-rank test). It was also found to be
significantly larger for the distal esophagus
(2.6[0.6]/7.3[1.2]/3.1[0.7]) and proximal stomach
(2.2[0.9]/6.8[1.1]/4.2[1.1]) than for the proximal
(1.4[0.4]/2.7[0.7]/1.3[0.4])
and
middle
(1.6[0.5]/3.2[0.6]/1.6[0.5]) esophagus in all three
directions (Fig. 1;
p
<0.05, Kruskal-Wallis with Dunn’s
test). Moreover, the SDs of peak-to-peak amplitudes and
marker positions at the inhalation and exhalation were
≤2.1mm (median: ≤0.9mm) in all three directions,
suggesting a small interfractional variability of the motion
amplitude and a stable trajectory shape (Fig. 2). Further,
a weak correlation (coefficient R: 0.54–0.71,
p
<0.001) was
found between the mean peak-to-peak amplitude and the
interfractional variability of amplitude and trajectory
shape (Fig. 2), implying that in addition to the peak-to-
peak amplitude, other factors such as stomach fillings
could also influence the interfractional variability of
amplitude and trajectory shape.
Conclusion
The amplitude and variability of the respiration-induced
esophageal tumor motion were found to be dependent on
direction and region. The limited interfractional
variability suggests that using a single planning 4D-CT may
be sufficient to take into account the respiration-induced
esophageal tumor motion.