730 / 1096
S714
ESTRO 36
_______________________________________________________________________________________________
prostatic bed (n=5, 29.4%) or prostate and local
recurrence (n=2 seminal vesicle, ischium 11.8%). Previous
treatment consisted on a median total dose of 74 Gy on
prostate or prostatic bed (range 66-76). Ten patients had
also received radiotherapy on seminal vesicles, four
patients on pelvic lymph-nodes. Median time from
previous radiotherapy was 80 months (range 26-116).
Median PSA at the moment of recurrence was 3.1 ng/ml
(average 4, range 1.2-13.5). As a re-irradiation, a median
total dose of 25 Gy (range 25-30) was delivered in a
median number of 5 fractions (range 5-6). An immediate
biochemical response was observed in all cases. Median
PSA nadir after treatment was 0.77ng/ml (average 1.33,
range 0.19-6.0, p=0.0004)) The sole acute toxicity
reported was genito-urinary, mainly represented by
pollakiuria and dysuria grad e 1 (n=9, 52.9%) or grade 2
(n=2, 11.8%). One patient (5.9%) had a grade 3 hematuria,
was hospitalized and submitted to continuous bladder
irrigation. A late grade 1 GU toxicity was observed in 3
patients (17.7%). No other toxicities were observed. At a
median follow-up of 16 months (range 6-36, calculated
from the time of recurrence diagnosis) 8 patients (47.1%),
experienced a biochemical recurrence, confirmed by a
positive PET-choline in 5 cases (29.4%). Median BFS was 19
months, 1- and 2-year BFS was 84.6% and 32.2%,
respectively. Median LC was 24 months, 1- and 2-year LC
was 90.9% and 40.4%, respectively. All patients are still
alive, 5 of them with measurable disease. Median OS was
96 months from the initial diagnosis (range 59-151).
Conclusion
With the technological novelties offered by modern
radiotherapy, re-irradiation of patients affected by
prostate cancer, and previously treated with radiation
therapy, confirms its safety and efficacy. Therefore, it can
be considered a valuable option for local recurrence of
this disease.
EP-1332 Contouring variability with CT and MRI of
prostate cancer for radiation planning
A. Otero-Romero
1
, A. Pérez-Rozos
1
, R. Correa-Generoso
1
,
I. Jerez-Sainz
1
, M.J. García-Anaya
1
, I. Zapata-Martínez
1
,
A. Román-Jobacho
1
, M.D. Toledo-Serrano
1
, R. Ordoñez-
Marmolejo
1
, I. García-Ríos
1
, J. Goméz-Millan
1
, J.A.
Villalobos-Martín
2
, T. Díaz-Antonio
2
, J.A. Medina-
Carmona
1
1
Hospital Virgen de la Victoria, Radiation Oncology,
Málaga, Spain
2
Hospital Virgen de la Victoria, Radiology, Málaga, Spain
Purpose or Objective
CT (Computer Tomography) is the standard for conformal
radiotherapy treatment planning of prostate cancer,
however T2-weighed MRI (Magnetic Resonance) allows
better definition of apex of prostate, seminal vesicles and
the rectum-prostate interface.
Analyse intra and inter-observer variability and whether
implementing systematic image fusion with CT and MRI
could improve prostate contouring accuracy.
Material and Methods
MR was requested to complete tumour staging and
performed in a different centre due to the unavailability
of MRI scan in our hospital. Planning CT was carried out in
our department, slices of 3 mm, with empty bladder and
rectum, in supine position using knee and feet
immobilization devices. Image fusion was performed with
T2-weighed MRI and CT scans matching on bony structures
of the pelvis.
We conducted the study in two parts.
First part of the study consisted in contouring the prostate
and seminal vesicles of a single patient on CT images and
then on MRI fusion images by 9 Radiation Oncologists
(including training doctors)
In the second part of the study two Radiation Oncologists,
specialized in prostate cancer, and a Radiologist trained
in MRI contoured the prostate of 5 patients on CT images
and then on MRI fusion images. The contour of the
Radiologist was considered the gold standard.
Comparison of volumes measured on CT and MRI using
Pinnacle planning system was made. Intraobserver and
interobserver variability was assessed taking into account
the percentage of coincident volume with the gold
standard, analysing the distance of the direction with
more differences, and calculating sensitivity (S) and
Paccard indexes (I
paccard
;P=delineated prostate; C=gold
standard).
Results
Accurate CT-MRI image fusion was not always achieved
with bony matching due to the different pelvis position
and
needed
soft
tissue
correction.
Volumes of the first part of the study range was 29.1-52.4
cc for prostate and 10.8-16.7 cc for seminal vesicles on
CT, and 29.5-57.2 cc for prostate and 11.6-16.1 cc for
seminal vesicles on MRI. Comparing CT and MRI volumes
the intraobserver ratio was 1.13 (1.02-1.26) for prostate
and 1.12 (1.01-1.21) for seminal vesicles.
In the second part of the study mean volumes range on CT
scan was 13-21 cm3 while on MRI was 18-26 cm3. Mean
volume% comparing to the gold standard volume range was
62%-67% on CT and 81%-86% on MR. Variability in distance
in the different directions were 3-9 mm in the longitudinal
axis, 3-4 mm in the lateral axis and 2-3 mm in the anterior-
posterior axis. Mean sensitivity index was 0.58 on CT and
0.80 on MRI, and mean Paccard index was 0.48 and 0.76
on CT and MRI respectivel
y.
Conclusion
Prostate MRI enables more accurate planning contouring
than CT. In our study CT volumes tend to be smaller than
on MRI. The longitudinal axis is the direction where more
contouring
differences
have
been
found.
MRI and CT could be made in the same pelvis position to
achieve reduced uncertainty image registration.
EP-1333 Impact of 18F-Choline PET scan acquisition
time on delineation of GTV in Prostate cancer
C. Parkinson
1
, J. Chan
2
, I. Syndikus
2
, C. Marshall
3
, J.
Staffurth
4
, E. Spezi
1
1
Cardiff University, School of Engineering, Cardiff,
United Kingdom
2
Clatterbridge Cancer Centre, Clinical Oncology,
Liverpool, United Kingdom
3
Cardiff and Vale University Health Trust, Wales
Research & Diagnostic PET Imaging Centre, Cardiff,
United Kingdom
4
Velindre Cancer Centre, Clinical Radiotherapy Trials,
Cardiff, United Kingdom