883 / 1096
S867
ESTRO 36
_______________________________________________________________________________________________
along SOBP dose profiles were predicted also for depth
positions where experimental data were not available. A
formula was also derived to predict cell death and
chromosome damage for a different cell line exposed to a
given ion type and energy, basing on the response of a
reference cell line to the same radiation quality. For both
endpoints, the increase of effectiveness along the plateau
was quantified. A non-negligible increase was found also
for protons, associated to high levels of damage beyond
the distal dose fall-off, due to the lower energy and thus
the higher biological effectiveness.
Conclusion
In line with other studies, this work suggests that assuming
a constant RBE along a proton SOBP may be sub-optimal.
More generally, this work represents an example of
therapeutic beam characterization avoiding the use of
experimental RBE values, which can be source of
uncertainties.
Acknowledgements:
this work was partially supported by
INFN (project ETHICS, P.I. L. Manti, local P.I. F. Ballarini;
MC-INFN/FLUKA, P.I. P. Sala, local P.I. A. Fontana)
EP-1607 Secondary cancer risk after particle therapy
for organs distal or lateral to the target volume
L. Toussaint
1
, L. Muren
1
, G. Engeseth
2
, C. Stokkevåg
2
1
Aarhus University Hospital, Medical Physics, Aarhus C,
Denmark
2
Haukeland University Hospital, Department of Oncology
and Medical Physics, Bergen, Norway
Purpose or Objective
Proton therapy is the most used particle therapy modality,
but carbon ions are also increasingly being applied for
specific tumour entities. Particle therapy in general has a
known potential of reducing the irradiated volumes of
normal tissues, although protons and carbon ions have
distinctively different dose distribution characteristics.
Protons have a steeper dose fall-off distally while carbon
ions have a sharper lateral dose penumbra. In addition,
carbon ions have a higher biological effect due to
increased cell inactivation, but also for the end-point cell
mutation associated with carcinogenic potential. The aim
of this study was therefore to compare the risk of
secondary cancer (SC) from dose distributions in the
thyroid and lungs, particularly radiosensitive organs
located distally and laterally to the target volume during
craniospinal irradiation (CSI). Since pre-clinical data
indicates that the carbon ions RBE for cell mutation may
be higher than for cell inactivation, we included this in the
models.
Material and Methods
CSI treatment plans with a prescribed dose of 23.4Gy(RBE)
were generated on CT-scans from six pediatric patients
(Syngo, Siemens) using pencil beam scanning protons
(IMPT) and carbon ions (C-ions). Relative risks (RRs) of
radiation induced cancer (IMPT/C-ions) for the thyroid and
the lungs were analysed by applying a bell-shaped dose-
response model (J Radiol Prot 2009; 29(2A): A143-157).
The model accounts for RBE, fractionation as well as for
the competing events of cell mutation and inactivation.
The RBE variation for the different end-points was
included by introducing and varying the ratio (k) between
cell mutation and inactivation for C-ions. The median and
range of the patient-specific RRs were calculated from the
physical dose distributions and the published input model
parameters.
Results
The dose distributions (Fig 1) illustrated the sharper
lateral penumbra of C-ions, which resulted in lower lung
doses compared to protons, while the C-ion fragmentation
tail contributed to higher doses to the thyroid than from
protons. The SC risk estimates strongly depended on the
ratio k, and the RR decreased for increasing k for both
organs (Fig 2). For the thyroid, the RR was higher from the
C-ion plans for the entire scanned range of k. Despite a
better sparing of the lungs with C-ions, the carcinogenic
potential of C-ions was not consistently lower than for
protons: Not including a difference in end-point resulted
in RRs in favour of C-ions, while increasing the ratio k gave
higher risks for C-ions compared to protons. For the lungs,
the median RR turned in favour of IMPT at a threshold
value
k=1.1.