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

21 / 146

RT dose with unchanged biological effect on tumor and potentially reduced
toxicity vs. standard RT (7 weeks, tumor: 70 Gy; margin: 63 Gy; lymph
nodes: 58 Gy). Our aim was to assess the feasibility, toxicity, and re-
sponse rate of the accelerated modulated fractioning.
Methods: The literature shows that higher MoAb-EGFr concentrations cor-
respond to steeper tumor cell survival curves. RB parameters were derived
from it and included in the model to obtain the daily dose to be delivered
to each target volume. To date, 2 of the 10 expected patients (patient 1 oro-
pharyngeal; patient 2 supraglottic squamous cell carcinoma) have been
recruited in the FAMOSO protocol for curative intent.
Results: The RB model suggested 6 weeks with daily increasing dose/
fractions: tumor 1.70, 1.95, 2.15, 2.30, 2.35 Gy; margins 1.50, 1.75, 1.95, 2.05,
2.10 Gy; lymph nodes 1.40, 1.60, 1.80, 1.90, 1.95 Gy. Both patients con-
cluded the RT treatment (patient 1 planned schedule; patient 2 with
interruption of MoAbEGFr, last 10 days with standard RT fractions), with
total dose to tumor of 62.7 and 61.8 Gy, respectively. Maximum acute tox-
icity (skin, mucosa) was G3. Follow-up is 6 and 2 months (patient 1: partial
response; patient 2 still under evaluation).
Conclusion: New treatment strategies combining RT with radiosensitizing
drugs, even accelerated, are feasible. The RB model is adequate provided
RB parameters are available from clinical data. Preliminary data from
FAMOSO are encouraging, suggesting feasibility and acceptable toxicity.
Longer follow-up and more patients studied are needed to confirm toxic-
ity findings and assess response rate.
A.52
INTEGRATION OF FMRI AND MEG FUNCTIONAL MAPS INTO A
CYBERKNIFE PLANNING SYSTEM: FEASIBILITY STUDY FOR MOTOR
ACTIVITY LOCALIZATION AND DOSE PLANNING
D. Cusumano * E. De Martin D. Duran F. Ghielmetti E. Visani
D. Rossi Sebastiano D. Aquin M. Marchett M.B. Bruzzone F. Panzic
L. Farisell a Postgraduate School in Medical Physics, University of Milan, Milan,
Italy; b Health Department, Foundation Carlo Besta Neurological Institute IRCCS,
Milan, Italy; c Department of Neurophysiology, Foundation Carlo Besta
Neurological Institute IRCCS, Milan, Italy; d Department of Neuroradiology,
Foundation Carlo Besta Neurological Institute IRCCS, Milan, Italy; e Department
of Neurosurgery, Radiotherapy Unit, Foundation Carlo Besta Neurological
Institute IRCCS, Milan, Italy
Introduction: Magnetoencephalography (MEG) and functional magnetic
resonance imaging (fMRI) provide non-invasive localization of eloquent brain
areas for pre-surgical planning. This study aimed to develop integration
MEG and fMRI maps into a Cyberknife (CK) system to optimize dose
planning.
Material and Methods: A patient with a brain metastasis affecting left the
pre and post-central gyrus underwent functional imaging of the hand motor
cortex prior to radiosurgery. MEG data were acquired with a 306 sensors
systemwhile the patient performed self-paced motor activation of the right
hand and index finger. Epochs were extracted in the window ranging from
– 3 to +3 seconds with respect to the movement onset and then averaged.
Motor activation of the right hand and index finger was obtained for fMRI
through a block designed paradigm. Stimulation modality and duration both
for MEG and fMRI were chosen to maximize time course signal to noise
ratio. MEG and fMRI maps were integrated into a CK system for treat-
ment planning optimization.
Results: Localization of the hand motor cortex was obtained for both
methods within proximity of the lesion. Integration of the fMRI data into
the CK system was easily achieved through the customary import proto-
col. More problematic was the integration of MEG images, and a customized
Dicom import software had to be developed. Only small differences were
observed between MEG and fMRI activations after image co-registration.
Inclusion of the activation area into plan optimization allowed a reduc-
tion of 19% of the mean dose to the motor cortex.
Conclusions: The availability of advanced neuroimaging techniques is
playing a more and more important role in radiosurgical planning strate-
gy. The authors developed an effective method to co-register fMRI and
MEG data sets in a CK system. This additional information can improve
dose sparing of eloquent areas, and MEG information in particular
might be valuable when BOLD effect is disturbed by pathological
vascularization.
A.53
SUM SIGNAL FILM DOSIMETRY: A NOVEL APPROACH FOR HIGH DOSE
PATIENT SPECIFIC QUALITY ASSURANCE WITH GAFCHROMIC EBT3
D. Cusumano * M. Fumagalli F. Ghielmetti L. Rossi G. Grossi
R. Lanzarotti L. Fariselli E. De Martin a School of Medical Physics,
University of Milan, Milan, Italy; b Health Department, Foundation Carlo Besta
Neurological Institute IRCCS, Milan, Italy; c Department of Computer Science,
University of Milan, Milan, Italy; d Department of Neurosurgery, Radiotherapy
Unit, Foundation Carlo Besta Neurological Institute IRCCS, Milan, Italy
Introduction: Film dosimetry with Gafchromic EBT3 is a widespread tool
to assess dose distributions in high precision radiotherapy. Its use becomes
difficult in radiosurgery (RS) and hypofractionated radiotherapy (HRT) due
to the limited film sensitivity at the high doses delivered with these mo-
dalities. The aim of this work is to propose and validate a novel method,
sum signal (SS), able to increase film sensitivity and effectively extend EBT3
use also to high dose patient specific quality assurance (QA).
Material and Methods: To characterize film dose response, 3 × 3 cm 2 films
were irradiated in the range of 0–24 Gy with a Cyberknife (CK), contextually
measuring delivered doses with an ionization chamber. Films were digi-
tized to obtain net optical densities (netOD) for red channel (RC) and green
channel (GC).
The SS was calculated by linearly summing the netOD values obtained for
RC and GC.
Calibration data were fitted using the effective variance method, taking into
account the uncertainties on netOD and dose values.
A homemade Matlab software was contextually developed to compare, in
terms of 3%/1 mm and 4%/1 mm gamma analyses, dose distributions de-
livered by CK on EBT3 to those calculated by TPS.
SS and single channel (SC) method (which consists of using RC for
doses < 10 Gy and GC for higher doses) were compared for 20 patient plans
and differences were evaluated by Wilcoxon signed-rank test
Results: Our analysis shows that the SS dose–response curve is character-
ized by a steeper trend in comparison with SC one, translating into higher
dose resolution over the whole dose range considered.
The impact of this increase was confirmed by gamma analysis, which shows
that the % of gamma points satisfying agreement criteria is significantly
higher for the SS method with respect to the SC: 94.1% vs 86.8% (p = 0.003)
for 3%/1 mm and 96.5% vs 92.7% (p = 0.004) for 4%/1 mm respectively
Conclusion: The results demonstrate that SS can be considered a novel and
effectivemethod able to improve dosimetric accuracy for RS and HRT patient
specific QA.
A.54
EXRADIN A26 MICROCHAMBER CHARACTERIZATION FOR SMALL FIELD
DOSIMETRY
C. Cutaia *, L. Radici, E. Gino, M. Pasquino, M. Stasi. Medical Physics
Department, A.O. Ordine Mauriziano, Torino, Italy
Introduction: The dose accuracy required in IMRT delivery quality assur-
ance is difficult to achieve when small radiation fields are generated. This
study aims to investigate the main dosimetric characteristics and the per-
formance in small photon beam dosimetry of one relatively new small field
detector.
Material andmethods: We tested an Exradin A26microionization chamber.
Short term stability, dose linearity, polarity effect, angular dependence and
energy dependencewere evaluated for field sizes up to 10× 10 cm2. The results
were compared with those obtained with an Exradin A1SL IC. PDD curves
and OAR profiles were acquired for 1 × 1, 2 × 2 and 3 × 3 cm2 field sizes; OF
values were measured for field sizes ranging from 0.6 × 0.6 to 10 × 10 cm2.
The results were compared with those obtained with an IC and an EDGE
diode.
Results: A26 IC readings showed 0.2% and 0.1% relative standard devia-
tions for 3× 3 and 10 × 10 cm2 fields, respectively. Dose response was linear
e16 Abstracts/Physica Medica 32 (2016) e1–e70