388 / 1082
S373
ESTRO 36 2017
_______________________________________________________________________________________________
validation) were 0.63, 0.68 and 0.66 respectively. Based
on this model, 2 groups risk were identified. The 3-year
OS, DFS and LRC were 88% (95CI: 67.4 – 100%), 78.7 % (63.6
– 97.3%) and 83.6% (95CI: 70.1 – 99.7%) for low-risk group
vs 45.5% (95CI: 23.8 – 86.8%), 33.3% (95CI: 15 – 74.2%) and
38.1% (95CI: 17.9 – 81.1%) for high-risk group (p<0.01)
(Figure 1 and 2).
Conclusion
TLG of the primary tumor and the distance between lymph
node and the primary tumor, weighted by the MTV of
lymph nodes, were correlated with LRC, DFS and OS.
These parameters seem to have a higher predictive value
than classical prognostics parameters, and may be useful
to tailoring the therapeutic approach in this type of
cancer.
PO-0719 The use of ultrasound bladder scanning in
cervical IMRT to reduce variability of uterine motion
S. Otter
1
, M. Hussein
1
, S. Why
1
, A. Franklin
1
, A. Stewart
1
1
Royal Surrey County Hospital, St Luke's Cancer Centre-
Oncology, Guildford, United Kingdom
Purpose or Objective
IMRT is being increasingly used over 3D conformal
radiotherapy in the treatment of locally advanced cervical
cancer. However, the uterus and cervix are mobile
structures whose positions vary depending on bladder and
rectal filling. The purpose of this study was to assess
whether the use of a portable US bladder scanning
protocol prior to each fraction of IMRT can reduce bladder
filling variability and therefore ultimately reduce cervical
/ uterine movement.
Material and Methods
Patients with locally advanced cervical cancer treated
with IMRT received daily CBCT imaging. The bladder was
retrospectively contoured on each CBCT. Ten patients
were treated using an US bladder scanning protocol (BVI
9400 bladder scan, Verathon). They had a daily pre-
treatment bladder scan (BS) and were imaged with CBCT
and then irradiated when the bladder volume fell in a pre-
determined individualized range based on planning CT
scans with empty and full bladder. If the bladder volume
was below the target range, patients were rescanned after
15 minutes, if it was above the range, they emptied their
bladder and recommenced bladder filling. Nine patients
were treated with a standard bladder filling protocol
which involved drinking 450ml water in 5 minutes then
treatment after 30 minutes. Results were analysed with a
Bland-Altman Plot to assess correlation between bladder
volume on US and on CBCT. Mann Whitney 1-tailed test
was used to assess whether the bladder scanning protocol
improved bladder filling consistency (the percentage of
radiotherapy fractions where the bladder volume on CBCT
fell within the pre-determined range).
Results
CBCTs and BS readings were available for 246 fractions of
radiotherapy in 10 patients in the BS cohort. 249 CBCTs
were available in 9 patients in the pre-BS cohort. The
mean difference between BS readings and bladder volume
on CBCT was -53.8cc, however the 95% CI band was wide
(38.6cc - 146.3cc) and there were a lot of outliers
suggesting a lack of statistical significance. This may be
related to a combination of: interoperator variability,
limited probe size, and differences in the time between
the BS readings and the CBCT. The percentage of fractions
that had bladder volumes within range were significantly
higher in the BS cohort (mean 70.5%) than the pre-BS
cohort (mean 38.5%, p=0.024). Two patients were
excluded from this analysis in the BS cohort as they had
empty bladder volumes greater than full bladder volumes
on planning CT.
Conclusion
The introduction of a bladder scanning protocol has
significantly improved the consistency of bladder filling
for each fraction of radiotherapy compared to a pre-
bladder scan cohort. This may enable a reduction in the
CTV-PTV margin in the future which would reduce the
volume of irradiated tissue and therefore hopefully reduce
toxicity.
PO-0720 CTV change during adaptive EBRT for cervix
cancer: is mid-treatment plan adaptation required?
D.D. Vignarajah
1
, M.C. Marbán
2
, K.T. Bath
3
, P.J. Hoskin
1
1
Mount Vernon Cancer Centre, Clinical Oncology,
Northwood, United Kingdom
2
Hospital Universitario Doctor Negrin, Radiation
Oncology, Las Palmas de Gran Canaria, Spain
3
Mount Vernon Cancer Centre, Radiotherapy Physics,
Northwood, United Kingdom
Purpose or Objective
Evaluation of clinical target volume (CTV) motion and
volumetric changes during cervical cancer external beam
radiotherapy (EBRT), to determine the value of mid-
treatment plan adaptation compared to pretreatment
plan adaptive radiotherapy.
Material and Methods
Patient Population
Patients undergoing pretreatment adaptive EBRT for
cervix cancer were eligible. Consistent bladder and bowel
preparation was used as per departmental protocol.
Analysis
Daily cone beam CT (CBCT) images were reviewed, with
the CTV and organs at risk (OARs) delineated on each
image set based on GEC-ESTRO guidelines.
Organ motion was assessed in 4 planes (anteriorly,
posteriorly, superiorly and inferiorly) at the same
anatomical points on each CBCT. Daily CTV volumes were
combined to form a weekly internal target volume (ITV),
with volume and motion assessed on a weekly basis. Based
on the changes noted, new PTV margins were added to the