ESTRO 36 Abstract Book

560 / 1082

S545 ESTRO 36 2017
_______________________________________________________________________________________________
Purpose or Objective
To further test the positioning accuracy of an optical
surface monitoring device called Align RT for the
treatment of breast cancer patients. The data will then be
analyzed to determine if the daily positioning with Align
RT is accurate enough to allow for fewer weekly MV
imaging of breast radiotherapy patients.
Material and Methods
16 breast cancer patients were treated using 3D tangential
fields. Patients were positioned supine on the breast
board using an optical surface detector (Align RT). After
positioning, MV imaging and bone match on the chest wall
was done to verify the patients position, and corrected
accordingly. All shifted values were recorded. The Align
RT system consists of 6 cameras, which acquire the
patients’ position in 2D, and a computer vision algorithm,
which reconstructs the image into 3D. The patients’
reference surface was imported from the CT scan and the
region of interest of the treated area was selected. The
patient was positioned by using the Monitoring mode in
Align RT and driving couch values until pre-shift Align RT
deltas were as close to zero as possible. The Vertical
(VRT), Lateral (LAT), Longitudinal (LNG) shift values
generated from the MV images equal to or less than 5mm
were marked as falling within our accepted tolerance for
breast patients. This data was further analyzed to
conclude if it would be acceptably accurate enough to
reduce daily imaging.
Results
Out of the 16 patients involved in the study, a total of 213
fractions were treated using Align RT. Of these fractions,
201 (94.4%) had MV shift values, in all directions, within
our 5mm tolerance. 209 of 213 fractions (98%) were within
a 6mm tolerance or less. The largest shift observed was
9.5mm in the lateral direction, however the most frequent
axis that fell out of tolerance were VRT and LNG. The most
failed fractions (fractions with shift greater than 5mm)
observed from an individual patient was 3 out of her 25
recorded fractions. 9 of the 16 patients were treated with
all fractions within tolerance.
Conclusion
The data was found to be very consistent across all
patients, with 9 from 16 patients having all MV shifts equal
to or less than 5mm in all directions after using Align RT
for positioning and the remaining having minimal fractions
outside this tolerance. This data suggests a strong
argument for reducing daily imaging with breast patients
being positioned with Align RT. Reducing daily imaging to
2-3 times per week in combination with daily positioning
using Align RT would then be valuable in reduction of both
excess dose to the patient and treatment time on the
Linear accelerator.
PO-0997 Improving shoulder positioning in a 5-points
mask.
L. Mesch1, S. Hol1, G. D'Olieslager 2, C. Buijs1, D.
Washington1
1 Dr. Bernard Verbeeten Instituut, Department of
Radiation Oncology, Tilburg, The Netherlands
2 Dr. Bernard Verbeeten Instituut, Department of medical
Physics & Engineering, Tilburg, The Netherlands
Purpose or Objective
Immobilization and positioning are necessary in radiation
therapy of head-and-neck patients. A 5-points mask offers
stability, although large shifts of the shoulder (>1cm)
routinely can occur. A wrong positioned shoulder can
cause coverage loss or an increased critical structure dose.
The purpose of this study was to decrease the
shouldershifts by increasing the awareness of the RTT’s.
Material and Methods
In the period January 2014 till June 2016 head-and-neck
patients were immobilized daily with a 5-points mask
(2014/2015: CIVCO Medical solutions: Posicast®, 2016:
Orfit Industries: Efficast® Push Pin)/Wedges 50 & 70
(Cablon Medical)/Head supports (2014/2015: CIVCO
Medical solutions: Posifix® Supine Headrests, 2016 Orfit
Industries: Raycast® Head supports regular density with
lateral neck flaps).
For each year, 10 patients with head-and-neck cancer
were selected randomly.
All patients had daily CBCT imaging (Varian Medical
Systems). The position of both shoulders on the CBCT
relative to the mask was quantified in inferior direction in
Offline Review (Varian Medical Systems). Measurements
were performed on the coronal plane, 1.5cm dorsal of the
isocenter (Figure 1).
Results
Three hundred CBCT scans (600 shoulder positions) were
examined. Table 1 shows the shoulder shifts in inferior
direction. In 2014 23% of the shoulders was positioned
≥15mm inferior relative to the mask. At the end of 2014
the results were presented at our institute. Extra
awareness of the RTT’s resulted in a decrease of the
shouldershifts ≥15mm (7%). After a second presentation in
December 2015, the percentage of shifts ≥15mm
decreased (3.5% vs. 7%) in 2016. The percentage of
shifts 5mm ≤ x < 10mm increased relative to 2015 (34.5%
vs. 24.5%). This might be due to lack of experience with
the new type of masks.
Conclusion
A presentation and discussion session is a simple and
efficient way to create more awareness. Extra awareness
of the RTT’s improved the setup of the shoulders in a 5-
points mask, but the immobilization of the shoulders
demands attention. The reason for the increase in shifts
5mm ≤ x < 10mm will be further investigated.
PO-0998 Comparison of the accuracy of different
pillow for radiation therapy of head and neck cancer
C.Y. Lin1, W.H. Yu1, C.Y. Lin1, J.T.C. Chang1
1Chang Gung Memorial Hospital, Department of
Radiation Oncology, Taoyuan, Taiwan
Purpose or Objective
Random error occurring in daily treatment is always a
serious concern in fractionated radiotherapy (RT). Image
guided radiotherapy can partially solve some of these
problems. However, the curvature of the cervical spine is
still a problem. Change of spine curvature can easily result
in a deviation of more than 3mm. Therefore, we initiate
this study to evaluate the accuracy of different pillow