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

66 / 146

the relevance of a new reference depth for dose prescription, were also
evaluated.
Materials and Methods: Beams with six energies between 30 and 150 kV
(HVL range in mm:0.3 Al to 1 mm Cu) from Pantak Therapax SXT150
(Elimpex) and nine cone diameters from 1 to 15 cm were used.
For in-water PDDs PTW ion chambers (34045 Advanced Markus, 31018
MicroLion, 31014Pin Point), 60003 PTW diamond detector and EFD3G IBA
diode were used, plus PTW 31010 Semiflex ion chamber for 150 kV.
For low-energy OFs, PTW 23342 soft x-ray ion chamber, and for medium-
energy OFs, Semiflex and PinPoint were used (‘low’ and ‘medium’ as IAEA
TRS 398).
A dose-deviation (1%) distance-to-agreement (1 mm) analysis was used to
compare each detector to MC EGSnrc system with DOSRZnrccode (and
SpekCalc code for primary-beam spectra).
As to extrapolation, a surface linear fit and a complete-PDD 6-th degree
polynomial fit were compared.
Results: Advanced Markus shows the best PDD performance. OFs mea-
sured with other three chambers are accurate within 4% and the linear
extrapolation error is less than polynomial.
Dmax ranges from 0.0225 mm to 0.75 mm; the ICRU-skin (0.075 mm) dose
is higher than the surface dose and it is very close to Dmax to within a few
percent for the studied beams.
Conclusions: Specific detectors were identified for accurate PDD and OF
measurements of 30 to 150 kV beams.
Referring to skin instead of surface for dose prescription is justified because
lower depths have no clinical relevance; the skin is closer to Dmax, and
extrapolations are no more necessary when using appropriate skin
dosimeters.
A.206
A NEW MEASURING SYSTEM OF MAMMARY GLAND THICKNESS PRIOR
TO INTRAOPERATIVE RADIOTHERAPY WITH ELECTRON BEAMS (IOERT)
IMPROVES DOSIMETRIC TREATMENT ACCURACY
P. Scalchi A. Marchesin G. Scalc F. Miol S. Noulas C. Baiocch
P. Francescon a Azienda U.L.S.S. 6 – U.O. Fisica Sanitaria, Vicenza, Italy;
b Azienda U.L.S.S. 6 – U.O.Chirurgia, Vicenza, Italy; c Azienda U.L.S.S. 6 – U.O.
Radioterapia, Vicenza, Italy
Introduction: Due to the depth-dose achievable in IntraOperative Elec-
tron Radiation Therapy (IOERT) of the breast cancer, the measurement of
the mammary gland thickness is a critical issue, because it determines the
electron energy to be chosen.
Usually a needle and a ruler (‘needle method’ hereafter), or ultrasounds,
are used before docking.
A new system (MARK’S, PAF Biomedica) for accurate measurements of the
gland thickness under treatment conditions was developed in Vicenza
Hospital.
The aim of this work is to compare the needle method to the new one in
terms of surgeon–surgeon variability and dosimetric consequences.
Materials and Methods: A mobile linac (LIAC, S.I.T.) with electron beams
of 4 to 10 MeV (R50 ranging from 17 to 38 mm of water) was used.
The MARK’S system is a sterilizable manual pointer with integrated ruler.
After radioprotective disk positioning, two surgeons stitch the mammary
gland to prepare the PTV, then insert the end of the applicator, after ap-
plying a thin patch layer underneath to prevent target herniation as well
as bubble formation, and while the applicator is kept pressed, the pointer
is inserted allowing direct thickness measurements in treatment condi-
tions. Pointer-applicator coaxiality improves reproducibility.
14 consecutive patients were measured by two surgeons before IORT, first
by the needle method, and then by MARK’S. Five measurements points were
always taken at fixed positions over the PTV.
Results: The needle method systematically overestimates the PTV thick-
ness. The maximum difference between two methods was 20 mm (5.4 mm
on the average).
The surgeon–surgeon variability is 1 mm and 2 mm, for MARK’S and the
needle method, respectively. If the latter was applied, 11 erroneous energy
choices would occur, and 5 treatments would be canceled following ICRU71
(2004) and AAPM TG72 (2006).
Conclusions: In breast IOERT, MARK’S system was verified to overcome the
accuracy concerns of the traditional needle method.
A.207
EVALUATION OF CT-BASED FEATURES AND ADC VALUES TO ASSESS
TUMOR CONTROL ON CERVICAL LYMPH NODES TREATED WITH CHEMO-
RADIOTHERAPY
E. Scalco S. Marzi G. Sanguineti A. Farnet G. Rizz A. Vidiri
a Institute of Molecular Bioimaging and Physiology (IBFM), CNR, Segrate, Italy;
b Medical Physics Laboratory, Regina Elena National Cancer Institute, Roma,
Italy; c Department of Radiotherapy, Regina Elena National Cancer Institute,
Roma, Italy; d Radiology and Diagnostic Imaging Department, Regina Elena
National Cancer Institute, Roma, Italy
Introduction: The aim of the study was to investigate whether structural
and functional information about cervical lymph nodes (LNs) extracted from
pre-treatment CT, and diffusion-weighted imaging (DWI), predict tumor
control to chemo-radiotherapy (CRT) of head and neck squamous cell car-
cinoma (HNSCC).
Materials and Methods: 14 patients with pathologically confirmed HNSCC
treated with CRT were enrolled, who underwent DWI and CT at baseline.
DWIs were obtained using three b values (b = 0, 500, 800 s/mm2).
The entire volume of LNs was contoured by a dedicated HN radiologist on
DWIs, using T2-weighted images as a guide for the LNs location; each LN
was deformed on CT by an elastic registration method. The median value
of the ADC volumetric distribution was considered for the analysis, besides
the mean intensity and entropy extracted from the CT. After a median follow-
up time of 2 years, 8 LNs were classified as being controlled (without
evidence of disease during follow-up on MRI and PET-CT) and were labeled
as RC LNs; 7 LNs were classified as having regional failure (pathology proven
residual tumor at neck dissection after CRT or during the follow-up) and
were labeled as RF LNs. Spearman’s correlation between CT features and
ADC were evaluated. The Fisher’s linear discriminant analysis was used to
determine which variables best discriminate between RC and RF.
Results: RC LNs presented higher mean CT values and lower CT entropy
and ADC with respect to RF. Significant correlations were found between
median ADC and mean CT values (R = −0.59) and entropy (R = 0.61). The
accuracy of the entropy measured on CT and the median ADC was 93.3%
and 80%, respectively.
Conclusion: Our preliminary results show that RC LNs have, at baseline,
lower CT entropy and ADC, and a higher mean CT intensity, highlighting a
significant correlation between DWI and CT information and suggesting that
more homogeneous and dense LNs before treatment may better respond
to CRT, probably due to limited areas of necrosis and hypoxia.
A.208
BOWEL DOSE-VOLUME PREDICTORS OF DIARRHEA IN PATIENTS TREATED
WITH POST-PROSTATECTOMY WHOLE-PELVIS IMRT
C. Sini *, C. Fiorino, L. Perna, V. Sacco, R. Calandrino, C. Cozzarini. IRCCS –
San Raffaele Scientific Institute, Milano, Italy
Aim: Gastrointestinal (GI) toxicity is a common side-effect of whole pelvis
intensity modulated radiotherapy (WP-IMRT), but clear dose–volume re-
lationships are still lacking. The aim of this study was to identify predictors
of diarrhea in a cohort of chemo-naïve patients treated with WP-IMRT after
prostatectomy.
Material and Methods: Clinical/dosimetric data in 115 patients (pts) treated
with adjuvant or salvage intent in a single institute were analyzed (static
field IMRT: 19; VMAT: 55; Tomotherapy: 41).
Dose–volume/surface histograms (DVHs/DSHs) for intestinal loops and
sigmoid colon were calculated. The primary endpoint was a change from
baseline of the Inflammatory Bowel Disease questionnaire (IBDQ) scores,
used to assess the degree of GI symptoms. Patient-reported scores at base-
line, at RT mid-point and end were prospectively collected. The frequency
of liquid defecation (item 5) was considered as the most clinically rele-
vant symptom. Score variation between baseline and half/end RT (D-
e61 Abstracts/Physica Medica 32 (2016) e1–e70