533 / 1096
S518
ESTRO 36
_______________________________________________________________________________________________
Material and Methods
Thirty advanced cervical cancer patients treated with HDR
(
192
I source) intracavitary brachytherapy were selected.
Rotterdam applicators (tandem-ovoids) were used for
them. Patients pelvic CT scans were done twice; pre- and
post-treatment (about 30 min after dose delivery), with
applicator in situ. Flexiplan
®
(version 2.6, Isodose control,
the Netherlands) as a 3D treatment planning software was
used. Applicator reconstruction and organs delineation
were done by the same physicist /physician on both image
sets. Totally identical plans (dwell times/positions) were
applied to both image sets and DVH parameters were
recorded; planning aims: 80-90 Gy (EQD2) for D
90
of CTV
HR
and less than 85, 75, and 75 Gy for D
2cm³
of bladder,
rectum, and sigmoid, respectively.
RT-Structure files (in DICOM format) of the patients for
whom intra-fractional dose (D
2cm³
) variations were higher
than 5% were exported from planning system. Applicator-
organs distances along the active length of three
applicators were extracted by some in-house MATLAB
written codes. Source dwell times were extracted from
treatment planning report files (in xps format). A model
was design to propose new source dwell times to
compensate for the bladder wall to applicators walls
distances intra-fractional variations, considering the TG43
algorithm and inverse square law. Some dwell times
acceptance criteria were considered during modeling such
as: D
90
of CTV
HR
and CTV
IR
have not changed to be less than
85 Gy 70 Gy, respectively. New dwell times were applied
to the plans to test their influences on DVH parameters.
Also, the model was further optimized to reduce the
executing time by searching for the most impressive part
of the applicators lengths on bladder dose.
Results
For one third of the considered patients bladder dose
changes were higher than 5%. Mean ± SD of D
2cm³
intra-
fractional relative changes ((D
2cm³(before)
- D
2cm³(after)
)/
D
2cm³(before)
× 100) of these ten cases were 19.3 ± 18.0 %.
After correcting the plans these variations became 10.5 ±
14.5. More bladder dose correction would lead to a
significant decrease in dose to CTV
HR
and was
unjustifiable. Model runtime was about 3 minutes (Intel
corei7 laptop, RAM = 8 GB, CPU = 2 GHz).
Conclusion
A model was developed to correct the bladder dose to be
as similar as possible to the pre-treatment plan one. It is
a semi-online model that can be used in the routine
clinical workflow to reduce the GYN image-guided
adaptive brachytherapy uncertainties. The model can be
generalized to other organs at risk.
PO-0936 Dose effects of draining rectal gas in image-
guided brachytherapy for gynecological cancer
H. Takase
1
, N. Ii
2
, Y. Yamao
1
, T. Kawamura
2
, M. Naito
1
,
Y. Watanabe
2,3
, Y. Toyomasu
2
, A. Takada
2
, H. Tanaka
2
, T.
Yamada
1
, H. Maki
1
, H. Sakuma
4
, Y. Nomoto
5
1
Mie University Hospital, Department of Radiology, Tsu,
Japan
2
Mie University Hospital, Department of Radiation
Oncology, Tsu, Japan
3
Matsusaka central hospital, Department of Radiation
Oncology, Matsusaka, Japan
4
Mie University Graduate School of Medicine,
Department of Radiology, Tsu, Japan
5
Mie University Graduate School of Medicine,
Department of Radiation Oncology, Tsu, Japan
Purpose or Objective
To verify the usefulness of draining rectal gas in image-
guided high-dose-rate brachytherapy for gynecological
cancer, we quantified the dose delivered to the rectum
and urinary bladder with and without draining rectal gas.
Material and Methods
From October 2013 to July 2014, 116 brachytherapy
fractions from 34 patients were performed in our
department for gynecological cancer. After the
applicators were inserted, cone beam computed
tomography (CT) images were obtained. Rectal gas was
observed in 11 brachytherapy fractions from 8 patients.
After draining rectal gas with a Nelaton catheter, cone
beam CT images were obtained again. Brachytherapy was
prescribed to point A using standard two-dimensional
dosimetry and planning. To quantify the dose delivered to
the rectum and urinary bladder, three-dimensional dose
distributions were calculated using the images from
before and after draining rectal gas. The dose to the
rectum and urinary bladder was evaluated based on dose-
volume histograms. The influence of the volume of
discharging rectal gas (pre-draining rectal volume – post-
draining rectal volume) on the rectal dose was
investigated. The minimum doses to the maximum
exposed 0.1, 1 and 2 cc (D
0.1cc
, D
1cc
and D
2cc
) volumes were
evaluated using the dose from point A. Statistical analyses
were conducted using the paired t-test and a linear
regression model.
Results
The mean rectal dose after draining rectal gas was
significantly lower than that before draining. The rectal
doses (D
0.1cc
, D
1cc
and D
2cc
) relative to point A at post-
draining vs. pre-draining were as follows: 106.9% vs.
121.2%, 88.3% vs. 98.6% and 81.5% vs. 90.9%, respectively
(p<0.05). The mean urinary bladder dose was not
significantly different after draining rectal gas from
before. The urinary bladder doses (D
0.1cc
, D
1cc
and D
2cc
)
relative to point A at post-draining vs. pre-draining were
as follows: 136.0% vs. 133.2%, 112.3% vs. 111.8% and
103.3% vs. 102.4%, respectively. The volume of
discharging rectal gas slightly correlated with the rectal
dose at D
0.1cc
(R2=0.45); however, no significant
correlation was found for D
1cc
or D
2cc
(R2=0.10 and -0.01,
respectively).
Conclusion
Our data suggested that draining rectal gas is useful for
reducing the rectal dose in high-dose-rate brachytherapy
for gynecological cancer.
PO-0937 HDR image-guided interstitial brachytherapy
for postoperative local recurrent uterine cancer
K. Yoshida
1
, H. Yamazaki
2
, T. Takenaka
2
, T. Kotsuma
3
, K.
Masui
2
, T. Komori
1
, T. Shimbo
1
, N. Yoshikawa
1
, H.
Yoshioka
1
, Y. Uesugi
4
, T. Hamada
1
, M. Nakata
1
, H.
Matsutani
1
, M.M. Ueda
3
, Y. Tsujimoto
3
, E. Tanaka
3
, Y.
Narumi
1
1
Osaka Medical College, Radiology, Takatsuki, Japan
2
Kyoto Prefectural University of Medicine, Radiology,
Kyoto, Japan
3
National Hospital Organization Osaka National Hospital,
Radiation Oncology, Osaka, Japan
4
Kansai University of Welfare Sciences, Rehabilitation
Sciences, Kashiwara, Japan
Purpose or Objective
In order to evaluate the usefulness of high-dose-rate (HDR)
image-guided interstitial brachytherapy (ISBT) for
postoperative local recurrent uterine cancer, we analyzed
our clinical experience.
Material and Methods
We investigated 48 patients treated with HDR-ISBT at
National Hospital Organization Osaka National Hospital
and Osaka Medical College between May 2003 and January
2014. All patients received radical surgery and 10 patients
also received post-operative radiotherapy as previous
treatments. Histologic finding was squamous cell
carcinoma (SCC), endometrioid adenocarcinoma (AD),
mucinous adenocarcinoma (MAD) and the others
(serous/adenosquamous/endocrine/undifferentiated) for
20, 17, 5 and 6 patients. The median maximum tumor
diameter was 25 mm (range; 5-79 mm). In 38 patients who
had non-irradiation history, 23 patients also received
external beam radiotherapy (EBRT). The median ISBT