197 / 1082
S184
ESTRO 36 2017
_______________________________________________________________________________________________
The uni- and multi-variate analysis showed statistical
significance for all parameters except lesion location
(table 1).
The high coefficient associated to the peripheral
character is probably due to the EPL un-modelled lateral
electron equilibrium caused by the prevalent presence of
low density lung tissues surrounding the peripheral lesion,
thus greatly impacting dose calculation differences. Based
on the multivariate analysis predictive nomogram was
generated (R²=0.58, Figure 1).
Dose to OARs calculated with EPL and MC showed strong
linear correlation (R²=0.99-1.00). The dose constraints
decreased by 1% in the heart (D10cc), great vessels
(D10cc) and spinal cord (D0.25cc), 2% in the oesophagus
(D5cc), 5% in the ribs (D5cc) and 16% in the trachea (D4cc).
.
Conclusion
The differences between MC and EPL are significantly
impacted by dose, tracking, location and the volume of
the lesion. Predictive nomogram helps to estimate the
differences on GTV D50. EPL to MC OAR dose tolerance
limit proved to have a strong linear correlation with
conversion factors ranging from 0.84 to 0.99. Based on our
model, re-prescription value can be estimated and, if
required, used to further restrict the constraints on the
OARs during EPL optimization.
OC-0350 ExacTrac®-based Fractionated Radiosurgery
(fSRS) of Choroidal Melanoma (CM)
M. Wösle
1
, P. Goldschmidt
1
, G. Lohm
1
, L. Grajewski
2
, L.
Krause
2
, I. Ciernik
1
1
Dessau City Hospital, Radiation Oncology, Dessau,
Germany
2
Dessau City Hospital, Ophthalmology, Dessau, Germany
Purpose or Objective
Proton therapy (PT) has been a standard for
treating choroidal melanomas for the last three
decades. However, PT is not easily available for the
majority of patients. Advances in photon therapy allow
highly conformal dose delivery while sparing normal
tissue. However, fSRS for small moving target volumes,
such as CM, has remained a challenge.
Material and Methods
Since December 2014, we treated 40 patients with central
choroidal tumors with fSRS. Radiotherapy plans were
obtained with iPlan® RT (Version 4.5.3, Brainlab,
Feldkirchen, D) after placement of 4 Tantalum clips
(Altomed Ltd., U.K.) a week prior the planning CT/MRT.
Thirteen cases have been reanalysed: ten treated with 10
Gy x 5 for CM, one case of haemangioma treated with 14,5
Gy in one fraction, and 2 cases of breast cancer metastasis
treated with 6 Gy x 5 and 4. iPlan® RT uses HybridArc™ as
field configuration with three to six dynamic conformal
arcs complemented with five to seven dynamic IMRT
fields. Treatment was delivered with a Novalis-
TrueBeam™ STx Linac (Palo Alto, CA, USA). IGRT was
achieved with ExacTrac® (version 6.0). ExacTrac®
documented the eye position before and after each
irradiation field.
Results
No local failure and no case of enucleation have been seen
at a median follow-up time of 329 days (range 98 –
678). Complications observed so far were impairment of
visual acuity in half of the cases, intraocular haemorrhage
(1 pts.), xerophthalmia (2 pts.), or keratritis (2 pts). The
minimal, mean, and maximal doses within the target
volumes were 86.3 (± 5.2)%, corresponding to V
95%
= 98.3
(± 1.4)%, 100.0 (± 0.4)%, and 103.4 (± 1.1)%, respectively.
The conformity index was 1.23 ± 0.16 and the homogeneity
index was 0.04 ± 0.01. Mean dose applied to the ipsilateral
eye lens, cornea, normal tissue of the involved eye, and
the lacrimal gland were 12.1 (± 17.7) Gy, 6.8 (± 11.5) Gy,
18.1 (± 9.2) Gy, and 10.7 (± 7.6) Gy, respectively. Maximal
dose of the ipsilateral optic nerve was 31.7 (± 15.9) Gy.
Doses delivered to the contralateral eye, lens, optic
nerve, and lacrimal gland were less than 0.6 (± 0.7) Gy,
0.2 (± 0.1) Gy, 1.3 (± 1.6) Gy, and 0.2 (± 0.1) Gy,
respectively. The doses given to the chiasma did not
exceed 1.3 (± 1.5) Gy; the pituitary gland 0.9 (± 1.4) Gy,
and the maximal dose to the brain did not exceed 11.1 (±
4.9) Gy. The ITV did not exceed 2mm.
Conclusion
HybridArc® in combination with peritumoral tissue
markers for image-guidance provides highly conformal and
homogenous doses distributions for the treatment of small
and moving target volumes.
Proffered Papers: Adaptive strategies
OC-0351 Analysis of concordance in multicentre
adaptive bladder trials quality assurance
E. Parsons
1
, D. Megias
1
, A. Baker
1
, S. Hafeez
2
, E. Hall
2
, H.
McNair
3
, Y. Tsang
1
, R. Huddart
3
1
Mount Vernon Hospital, National Radiotherapy Trials
Quality Assurance Group, Northwood- Middlesex, United
Kingdom
2
The Institute of Cancer Research, Clinical Trials Unit,
Sutton- Surrey, United Kingdom
3
Royal Marsden Hospital, Radiotherapy, Sutton- Surrey,
United Kingdom
Purpose or Objective
HYBRID (CRUK/12/055) and RAIDER (CRUK/14/016) are
two randomised phase II multi-centre clinical trials
investigating the use of adaptive 'Plan of the day” (POD)
bladder radiotherapy. In order to promote accurate POD
selection across multiple recruiting centres, a pre-accrual
assessment for adaptive plan selection was developed as
part of an IGRT QA credentialing programme (3
rd
ESTRO
Forum 2015-OC0564). The purpose of this study was to
establish whether the pre-accrual POD assessment is a
feasible QA process and investigate whether the POD can
be consistently selected by individuals across all recruiting
centres.
Material and Methods
Twelve bladder CT/CBCT image pairs were made available
to individuals to register according to protocol. Individuals
recorded the most appropriate POD selection from a
library of three possible plans and submitted to a central
QA group for review. In order to receive QA approval to
select the POD for HYBRID/RAIDER, individuals were
required to achieve an assessment score of ≥83% (10/12)
agreement with the expert consensus answers.
It was a pre-requisite of HYBRID and RAIDER that centres
already have an appropriate IGRT competency framework
for bladder CBCT in place prior to trial recruitment. The
assessment was first piloted for 10 centres recruiting to