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

22 / 146

to within 0.5% both for 3 × 3 and 10 × 10 cm2 fields. As dose-rate depen-
dence regards, the microchamber readings showed the maximum variation
from the reference condition at the lowest dose rate value; relative stan-
dard deviation of measurements was within 0.3%. The polarity effect is within
0.5% for field size greater than 2 × 2 cm2; for field size less than 2 × 2 cm2
the effect increases, still remaining within 1%. No energy and angular de-
pendence were observed. A26 IC PDD measurements agreed with those
obtained with the EDGE diode. FWHM values show a better agreement with
the EDGE diode values than A1SL. The penumbra values were greater than
those obtained with the EDGE diode. OF values showed a good agree-
ment with EDGE diode ones.
Conclusions: A26 IC meets the requirements to be a class reference chamber
and it could consequently represent a reference detector for PDD and OAR
measurements.
A.55
FIELD-IN-FIELD VERSUS 3D-DYNAMIC WEDGE TECHNIQUES FOR
PATIENTS WITH BREAST CANCER: A PRELIMINARY STUDY
P. D’Avenia * R. Nigro M. Camarda E. Di Nicola M. Giannini
M. Mangiacott F. Mascioni L. Montani S. Riccardi G. Rossi S. Fattori
a Medical Physics Unit, Macerata Hospital, Macerata, MC, Italy; b Radiotherapy
Unit, San Camillo De Lellis Hospital, Rieti, Italy; c Radiotherapy Unit, Macerata
Hospital, Macerata, MC, Italy
Introduction: The aim of the study was to compare the field-in-field (FIF)
and the 3D-CRT with dynamic wedge (DW) techniques for breast radio-
therapy in terms of dose distribution and reproducibility.
Materials and methods: Twelve patients with breast cancer fromMacerata
and Rieti Hospitals were studied (PTV range: 198–2407 cc). For each patient,
two plans with opposed tangential fields were prepared: tangential plan
with DW and tangential plan with FIF (Eclipse, Varian).
Two endpoints were analysed. The first endpoint was a dosimetric com-
parison between FIF technique with 6MV and 3D technique with DW (mixed
energies).
The second endpoint was to verify the reproducibility of plans for both tech-
niques: all patients repeated CT scan twice during the radiation treatment.
The original plan was recalculated on repeated scans registered with first
simulation CT, keeping the same MUs.
V95% (% PTV volume receiving 95% of prescription dose), D95% (dose in
Gy covering 95% of PTV volume), Dmean, homogeneity index (HI) and con-
formity index (CI) were used for comparison.
Results: Even if the sample is too small for a statistical analysis, some con-
siderations can be made. Regarding the comparison between FIF and DW,
V95%, D95%, and Dmean were higher for FIF technique than DW; HI was
lower in FIF technique, indicating a more homogeneous dose distribution
within PTV; CI value was between 1 and 2 in both techniques, indicating
a good adaptation of the reference isodose to the PTV.
Plans calculated on the repeated CTs showed smaller differences with the
original plan when using FIF technique. Comparison was made in terms
of V95%, D95%, Dmean, HI and CI.
Conclusions: In our preliminary study, FIF technique results in better dose
distribution in terms of homogeneity and planning reproducibility.
Further investigations involving an increased number of patients should
be performed to determine the impact of FIF technique also on organ at
risk toxicities.
A.56
AAA VERSUS PBC ALGORITHM FOR LUNG CANCER TREATMENT
PLANNING: DOSIMETRIC ANALYSIS FOR TARGET VOLUME AND ORGAN
AT RISK
P. D’Avenia M. Camarda E. Di Nicola M. Giannini F. Mascioni
L. Montani G. Rossi S. Fattor a Medical Physics Unit, Macerata Hospital,
Macerata, Italy; b Radiotherapy Unit, Macerata Hospital, Macerata, Italy
Introduction: Retrospective analysis of lung cancer 3D-CRT plans is made
to investigate dosimetric differences between PBC and AAA algorithm.
Materials and methods: The 3D dose distributions of 50 treatment plans
of lung cancer produced using PBC algorithm (Eclipse, Varian) were recal-
culated using AAA with the same monitor units (MU) number provided by
PBC and clinically delivered to each patient. The energy used for planning
was 6 MV. The differences in dose distributions and dose limits for organs
at risk were studied by comparing dosimetric parameters extracted from
DVH.
The Wilcoxon rank test was used for statistical comparison of all results
obtained from the use of the two algorithms: p value < 0.05 was consid-
ered the threshold for statistical significance.
Results: For the PTV, statistically significant difference was observed for
V95% (% PTV volume receiving 95% of prescription dose) and Dmax (%) on
PTV: in particular, the median values of these parameters were higher for
AAA plans than PBC plans (V95%: 89.3% vs 87.9% p = 0.009; Dmax: 107.8%
vs 106.9% p = 0.003). Instead, no statistically significant difference was ob-
served for D95% (dose in Gy covering 95% of PTV volume), conformity index
and homogeneity index.
For organs at risk, the median values for spinal cord, heart, esophagus and
V20 of total lungs (% lung volume receiving at least 20Gy) were found to
be higher for AAA plans, but always within the limits reported in the lit-
erature and the differences were statistically significant. No difference was
observed for Dmean of total lungs.
Conclusions: When the original PBC treatment plans were recalculated with
AAA, keeping the same MU number provided by PBC, the maximum dose
and coverage of PTV were higher, and the dose values to organs at risk in-
creased. As future work, the AAA plans will be optimized independently
from PBC plans and dosimetric differences will be statistically evaluated.
A.57
GEL DOSIMETERS FOR RADIOTHERAPY APPLICATIONS: RESULTS OF A
“RESEARCH PROJECT OF NATIONAL INTEREST” (PRIN)
F. D’Errico L. Lazzeri M. Marian D. Dond M. Marrale P. Randacci
G. D’Agostino a Università di Pisa, Pisa, Italy; b Politecnico di Milano, Milano,
Italy; c Università degli Studi di Pavia, Pavia, Italy; d Università degli Studi di
Palermo, Palermo, Italy; e Università degli Studi di Cagliari, Cagliari, Italy;
f Istituto Nazionale di Ricerca Metrologica, Pavia, Italy
Introduction: Gel dosimeters for three-dimensional mapping of radio-
therapy doses were introduced at Yale University in the mid-1980’s. Soon
after, research and development in this field also started in Italy. Early work
was done at the Istituto Superiore di Sanità and at the Universities of Pisa
and Milan. Several institutes now collaborate on this topic with the goal
of developing new formulations of hydrogel matrices with improved char-
acteristics of stability, sensitivity and spatial resolution compared to those
of earlier Fricke-gel and polymer-gel systems.
Materials and methods: Contrary to earlier gels based on natural gelling
agents, such as porcine skin gelatin and/or agarose, which suffer from limited
batch-to-batch reproducibility, we chose to develop formulations based on
synthetic polyvinyl alcohol chemically cross-linked with glutaraldehyde.
Several gel formulations and production procedures were investigated and
are illustrated. The properties of these gels were investigated through a
variety of techniques, including curemetry to determine the influence of
GTA concentration and temperature over gelation time as well as scan-
ning optical absorption to determine the sensitivity of the detectors and
the spatial stability of the response.
Results: Comparative results are presented for the formulations we inves-
tigated, including one that provides a lower detection limit on the order
of 100 mGy and a linear response exceeding 30 Gy, which makes it suit-
able for the verification of complex 3D treatment plans.
Conclusions: Our project has achieved the overall goal of developing 3D
dosimeters suitable for the verification of the treatment plans delivered
in modern radiotherapy techniques. Compared to earlier version of the de-
tectors, we also achieved a simplification of the manufacturing processes
in order to facilitate the diffusion of this technology and its transition from
the bench to the bedside.
e17 Abstracts/Physica Medica 32 (2016) e1–e70