Physica Medica: European Journal of Medical Physics - Volume 32 - Supplement 1 - February 2016

16 / 146

and 400 MeV/u) beams as well as to radiotherapy photon (6, 18 MV and
60 Co) and electron (6 and 12 MeV) beams. The measurements were per-
formed at ‘National Center for Oncological Hadrontherapy’ (CNAO Pavia,
Italy) for proton and carbon ion beams and at ‘San Raffaele Hospital’ (HSR
Milan, Italy) for photon and electron beams.
Results: The dose–response curves were fitted to a quadratic-like func-
tion with a relative error of less than 3%. The EBT3 response to protons was
found to be very similar to the response of photon and electron beams and
no saturation occurred. The relative response of carbon ions was less than
unity for all dose values and lower for low-energy ions. The carbon ion curves
show an under-response, up to 25% for the lower investigated energy, in
comparison with photon and electron curves.
Conclusions: For proton beams, at all energies, the dose–response curves
are statistically not different from photon and electron curves. For carbon
ions an under-response was observed which is higher for lower energy
(higher LET) beams.
A.35
SBRT PRE-TREATMENT QA
N. Cavalli N. Ricottone E. Bonanno G. Pisasal A. D’Agostino
A. Girlando C. Marin a School of Medical Physics, University of Catania,
Catania, Italy; b Humanitas – Centro Catanese di Oncologia, Catania, Italy
Introduction: In this study an IBA 3D dosimetry analysis package, COMPASS
3.0 with MatriXXEvolution ion chamber array, has been tested for SBRT pre-
treatment verifications in terms of 3D dose distribution, gamma analysis,
target and OAR structure DVH. The goal of the study has been to check the
capability of the system to detect possible errors of delivery and to compare
its results with our routinely pre-treatment verification system, based on
EPID images and EPIQA software.
Materials and methods: Considering a SBRT treatment on a liver disease,
delivered in VMAT mode, as reference plan, shifts on one single leaf have
been applied; the leaf has been retracted of 2 mm, 3 mm, 4 mm, 5 mm and
finally the same leaf has been completely locked into the field. In a second
step the plan was re-calculated by closing the X1 jaw of 1.5 mm, 3.5 mm,
4.5 mm and 6.5 mm.
Differences in D100% and D50% and local γ analysis (2 mm/2%–3 mm/3%)
were investigated for CTV, PTV and OARs. The same cases were analyzed
in terms of γ analysis (2 mm/2%–3 mm/3%) with Epid (and EPIQA).
Results: Using a 2%–2 mm criteria the first detectable error with EPIQA
regards a 3.5 mm X1 jaw shift and a 4.5 mm X1 jaw shift with COMPASS.
Comparable results about MLC misplacement have been found with the
two systems. For the 3%–3 mm criteria both systems show failing gamma
test from 6.5 mm X1 jaw shift and 4 mm MLC error positioning.
PTV coverage differences between wrong and correct plan show a differ-
ence of 1.4 Gy in D100%, for COMPASS reconstructed plan, regarding a
3.5 mm X1 jaw shift; similarly the first alarm bell in D100% difference is
observed for a 4 mm MLC misplacement.
Conclusions: Both devices have shown a high sensitivity in detecting jaws
and MLC movement. A 3%–3 mm criteria is not sensitive enough for the
two systems. The identification of systematic errors should be a very im-
portant goal in the pre-treatment process. This work confirms that gamma
approach for SBRT pre-treatment verifications could not be so crucial to
decide about delivery.
A.36
THE ROLE OF PINNACLE3 SCRIPTING IN STREAMLINING IMRT AND 3D-CRT
TREATMENT PLANNING
A. Chiappiniello *, C. Fulcheri, R. Tarducci. Medical Physics Unit, Santa Maria
della Misericordia Hospital, Perugia, Italy
Introduction: Pinnacle3 scripting is a powerful and versatile tool for treat-
ment planning, but, unfortunately, it is not officially supported by Philips
except for very specific uses. As Pinnacle3 scripting development and clin-
ical implementation requires resources, it is left to experienced Pinnacle3
users. The aim of this study was to assess the impact and role of Pinna-
cle3 scripting in planning routine.
Materials and methods: 7 Pinnacle3 scripts were developed, tested and
clinically validated. These scripts support the medical physicist in the beam
setting for prostate and head and neck IMRT plans, in the definition of ROIs
concerning the target for IMRT plans and the dose prescriptions on them,
as well as in the setting of beam boxes for 3D-CRT plans. Moreover, a script
that switches from relative isodose curve visualization to absolute visu-
alization (and vice versa) was implemented. The impact assessment was
performed evaluating planning times of several IMRT and 3D-CRT plans.
The development of the above-mentioned scripts took about 2 weeks for
a physicist with basic programming experience, whereas the training period
on using scripts for the medical physicists lasted up to 4 hours.
Results: Using the developed scripts a significant reduction in the elabo-
ration time of the treatment planning was observed (up to 70% for head
and neck IMRT plans with 3 different dose levels). Furthermore, other ad-
vantages of scripting are: the standardization of treatment planning (which
facilitates working with protocols), the adoption of a unique naming con-
vention for beams, points and ROIs, as well as the reduction of the operator-
dependent planning phases. For these reasons, planning with scripts is
generally less prone to human errors.
Conclusions: The authors recommend scripting for Pinnacle3 users in order
to standardize and streamline the treatment planning. Moreover, scripts
can be integrated with Pinnacle3 Auto-Planning in order to improve the au-
tomation of most planning phases.
A.37
CLEAR AND FILL IN MONACO 5.0: IS IT FEASIBLE TO DELIVER HIGH DOSE
WITH LOW MONITOR UNITS IN STEREOTACTIC BODY RADIATION
THERAPY (SBRT) FOR LUNG CANCER?
S. Ren Kaise M. Chieregato * M. Galell C. Bassetti M. Bignardi
a Fondazione Poliambulanza Istituto Ospedaliero, Brescia, Italy; b Scuola di
Specializzazione Fisica Medica, Università di Milano, Milano, Italy
Introduction: This study explores an optimization tool of Monaco 5.0 TPS
with the aim of sparing delivering time of the SBRT treatments of lung cancer
in order to meet the clinical request.
Materials and methods: A PET phantom (L981606, PTW) is filled with cork
grains to substitute lung tissue; six embedded spheres are water-filled to
simulate PTV and OARs. An air-filled cavity is made with a syringe placed
close to a sphere; an ionization chamber (PTW31010) can be inserted inside
the syringe. PTV contour encompasses the lesion sphere and the air-filled
syringe both surrounded by cork granules in order to simulate patient
anatomy. CTV includes the syringe and is drawn with a 3 mm margin from
PTV. CT scans of the handmade thorax phantom are imported in Monaco.
VMAT and Dynamic Arc (DynA) treatments are calculated with a single frac-
tion dose of 3 (CTV) and 2.7 (PTV) Gy. For all plans the same constraints
and calculation parameters are kept except for the application on PTV of
the Monaco tool “Clear and Fill” (C&F). Four plans are computed: 2 VMAT
and 2 DynA, with and without C&F tool. For all plans the total number of
monitor units (MU) is recorded. QA plans are also calculated in CT scans
with IC inserted and compared with measurements.
Results: VMAT (DynA) C&F plan delivers 75% (94%) of the dose with one
fifth (94%) of MU number of that without C&F. Dose differences com-
puted in QA plans are confirmed by measurements.
Conclusion: Cork physical density and relative electron density (RED) affect
lung dose primarily in small fields, typical of VMAT. Cork grains used in
our study have a RED of 0.09, which is quite far away from the 0.22–0.26
RED of human lung: the lower the densities, the bigger the dose and MU
difference. Our thorax phantom represents an extreme case of human lung
so dose differences found are large. Preliminary comparisons of clinical plans
with C&F tool show reduced dose discrepancies, prompting us to study again
the feasibility to deliver higher dose with lower MU.
e11 Abstracts/Physica Medica 32 (2016) e1–e70