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

73 / 146

Results: Average percentage volumes covered by the prescribed dose
were significantly different in RT and MCrecalc plans for different tumor
size for GTV (GTVsmall_V = −17.4 ± 18.8%, GTVlarge_V = −6.6 ± 7.4%, p < 0.05)
and PTV (PTVsmall_V = −51.9 ± 21.9%, PTVlarge_V = −23.7 ± 12.1%, p < 0.01).
For GTV, average Dmean and D50 differed between RT and MCrecalc
plans according to tumor size (small: Dmean = −11.8 ± 6.7%,
D50 = − 11.5± 6.6%; large: Dmean = −5.1 ± 2.1%, D50 = −5.0 ± 2.1%, p < 0.01).
Analogous behavior was identified for PTV (small: Dmean = −18.6 ± 9.6%,
D50 = −18.5 ± 10.7; large: Dmean = −7.0 ± 2.9%, D50 = −6.3 ± 2.9%, p < 0.01).
Differences for lungs in V20, V10, and V5 are −0.6 ± 0.4%, −1.4 ± 1.4% and
−3.4 ± 4.4%, respectively.
Conclusion: Both GTV Dmean and D50 could be used to evaluate target
dose underestimation for lung lesions. Based on GTV dose parameters dif-
ference between RT and MC calculations, a different correction to dose
prescription for small and large tumors could be adopted. The relation to
tumor position and beam angles geometry should be further investi-
gated. Target under-coverage suggests that optimization with MC algorithm
needs to be introduced in treatment planning.
A.229
A VERSATILE ASYMMETRIC COLLIMATION SYSTEM FOR CLINICAL USE IN
EXTERNAL ELECTRON BEAM RADIOTHERAPY WITH ELEKTA’S LINAC
A. Valentini L. Menegotti D. Ravanelli K. Najera a Medical Physics
Department, Azienda Provinciale per i Servizi Sanitari – APSS, Trento, Italy;
b The Abdus Salam International Centre for Theoretical Physics – ICTP, Trieste,
Italy; c Radiotherapy Department, Instituto de Cancerologia – Dr. Bernardo del
Valle S., Guatemala, CA, Guatemala
Introduction: In June 1996 the Medical Physics Department (MPD) of S.
Chiara Hospital in Trento designed and realized an electron collimator to
be adapted on a standard Elekta’s Linac cone applicator for electron beam
radiotherapy treatments. The major innovation was the variable asym-
metric rectangular collimation of the electron beam, not offered by the
vendors until today. Another outstanding feature is its interchangeability
between the two different Elekta’s Linacs installed at S. Chiara Hospital in
Trento used for electron therapy.
Materials and methods: Electron beams were used firstly for patient spe-
cific head and neck boost and now mostly for patient specific breast boost.
For such treatments the MPD had the clinical necessity to create an asym-
metric collimator and designed it since no vendors offered such products.
Brass knurled was used to build a variable asymmetrical collimator, to be
fixed to the standard squared electron applicator of 10 cm × 10 cm.
Results: The collimation of electron beams is obtained by moving two or-
thogonal knurled brass plates of 8.2 mm thickness, resulting in 50 different
field collimation shapes and sizes within squared electron applicator of
10 cm × 10 cm. From June 1996 up to now, approximately 6000 electron
treatments were delivered using this collimator, i.e. 80% of the electron treat-
ments performed at the Radiotherapy Department at S. Chiara Hospital in
Trento. The replacement of lead blocks shielding with the new asymmet-
ric collimation system for electron beams, leading to spare time and money,
improving the reproducibility.
Conclusions: Nowadays this electron collimator is still used in clinical elec-
tron treatments due to its ease of use and practicality. For 19 years there
were no problems related to its clinical use, showing high reliability and
mechanical robustness.
A.230
COMMISSIONING OF FLATTENING FILTER FREE BEAM FOR STEREOTACTIC
ABLATIVE RADIOTHERAPY (SABR) LUNG TREATMENTS
S. Valzano * M. Fusella E. Mones C. Secco F. Puricelli G. Loi
M. Brambilla a Medical Physics Unit, AO Ordine Mauriziano, Turin, Italy;
b Medical Physics Unit, AOU Maggiore della Carità, Novara, Italy
Introduction: To commission a 6 MV flattening filter free (FFF) photon beam
from a LINAC Varian Trilogy TX in the RayStation TPS.
Materials: In order to commission the FFF beam in a TPS, based on a three
source model, PDD, profiles and output factos (OF) were acquired in a water
phantom with isocentric setup. The following detectors were used: an IBA
CC13 chamber, an Exradin A16 μchamber, a Sun Nuclear edge diode (SE),
an IBA diode (SFD) and an Exradin W1 scintillator (ES). CC13 was used in
large field dosimetry, for small field sizes, ranging from 0.6 cm2 to 3 cm2;
all the other detectors were used and their outcomes compared in order
to study and minimize the impact of dose uncertainties on the beammodel.
The commissioning was validated by end to end tests in simple and complex
geometry in phantom. VMAT SABRT was planned using the FFF beam in
phantom and real patients and compared with the standard (FF) beam plans
by means of DVH analysis and patient specific QA passing rate obtained
with ArcCHECK.
Results: The SE diode performed better than A16 and SFD in small fields
with reduced partial volume effects and less energy dependence, then it
was our detector of choice for PDD and profile scans. The diode OFs were
in agreement within 2% with ES down to the field 0.8 cm 2. Only for the
smallest field the ED and SFD overestimate the ES by 11% and 7%
respectively; the ED reading was corrected on the basis of the other two
ones. The beam commissioning using these data met successfully the
Van Dyk criteria with minimal OF corrections between calculated and
measured data. End to end test showed good agreement between calcu-
lated and measured doses with passing rates > 90% for gamma(2/2)
criteria. No significant differences were found between FFF and FF plans
QA passing rates and in DVH analysis.
Conclusions: FFF beams can be modeled in RayStation with dosimetric per-
formances comparable with FF beams in lung SABR, but with relevant time
sparing and improved treatment accuracy.
A.231
STUDY OF THE DOSE DELIVERY SYSTEM INACCURACIES AND THEIR
IMPACT ON THE DOSE DISTRIBUTION DURING THE FIRST YEARS OF THE
CNAO CLINICAL ACTIVITY
A. Vignati M.A. Hosseini A. Attili M. Donetti S. Giordanengo
L.F. Guarachi F. Marchett F. Mas Milian A. Mirandola S. Molinelli
V. Monac G. Russo R. Sacch M. Varasteh Anva R. Cirio a Istituto
Nazionale di Fisica Nucleare, Torino, Italy; b Ionizing and Non-ionizing Radiation
Protection Research Center (INIRPRC), Shiraz University of Medical Sciences
(SUMS), Shiraz, Iran; c Centro Nazionale di Adroterapia Oncologica (CNAO),
Pavia, Italy; d Università degli studi di Torino, Torino, Italy; e CNPq Fellow,
Universidade Estadual de Santa Cruz, Bahia, Brazil; f Internet-Simulation
Evaluation Envision (I-SEE) s.r.l., Torino, Italy
Introduction:The Italian hadrontherapy center (CNAO) uses actively scanned
proton and carbon-ion beams to treat tumors and is equipped with a dose
delivery system (DDS) to monitor and guide the beams to the patient. This
work aims at evaluating the impact of the DDS inaccuracies on the dose
distribution of patients treated at CNAO through a retrospective analysis
of the data collected during the delivery (D) and their comparison with the
planning (P).
Materials and methods: The DDS delivers the dose in spots according to
the P, each spot being defined by the number of particles, the beam posi-
tion and energy. The same quantities are measured by the DDS during the
D and saved in the LOG files. The absolute difference between D- and P-spot
positions at the reference plane at the isocenter, and the relative differ-
ence between the P- and D-number of particles were assessed. Then, the
P- and D-spot quantities were used as inputs of a Forward Planning program,
which computes the dose of a treatment configuration, based on the patient
anatomy and the beams set-up. The obtained P- and D-dose maps were
compared through the Gamma-index, which measures the discrepancy
between two dose distributions, in terms of spatial distance and dose
difference.
Results: 61 patients (01/2012–04/2013) were studied. 98% of the spots
showed position deviations less than 1.5 mm, and differences in number
of particles were less than 2.5%. The P- and D-dose maps were compared
for 5 patients, representing time periods with different accuracy level of
the DDS. Even in the worst case, more than 98% of points passed the (3 mm,
3%) Gamma-index criteria.
Conclusions: The accuracy of the CNAO DDS was evaluated in terms of de-
viation of the spot position and number of particles between P- and D-spots.
Variations in the position accuracy were correlated with fine tunings of the
e68 Abstracts/Physica Medica 32 (2016) e1–e70