ESTRO 36 Abstract Book

105 / 1082

S92 ESTRO 36 2017
_______________________________________________________________________________________________
decrease the risk of second IBTR with acceptable cosmetic
results and low rate of late side effects.
Poster Viewing : Session 4: Brachytherapy
miscellaneous
PV-0183 Microbrachytherapy: even more localised dose
profiles?
R. Brown 1,2, X. Franceries1,2, M. Bardiès1,2
1 INSERM, UMR1037 CRCT- F-31000, Toulouse, France
2 Université Paul Sabatier, UMR1037 CRCT- F-31000,
Toulouse, France
Purpose or Objective
Owing to its intrinsic ability to deliver increased dose rates
to tumours whilst respecting organ at risk (OAR)
constraints, brachytherapy (BT) is being increasingly used
for the treatment of radioresistant tumours.
A new form of BT, microbrachytherapy (MBT), is proposed
for small tumours. For this treatment, the grains used in
BT are replaced by a solution containing β-emitters. More
injections can be used with MBT than grains with BT,
allowing for greater precision when targeting the tumour.
As with all forms of radiotherapy, treatment planning is
required. In this work, a method of generating optimal
MBT treatment plans is proposed.
Material and Methods
The non-dominated sorting genetic algorithm II (NSGA2)
[1] is used to generate treatment plans. This is a multi-
objective algorithm, permitting the objective functions to
be optimised independently.
Two objective functions were used: the first to minimise
the fraction of the tumour receiving less than the target
absorbed dose (60 Gy) and the second to minimise the
number of injections.
The algorithm was validated on a spherical tumour of
20 mm radius. 20 mm was chosen because it represents
the typical size of tumour that could be targeted with this
new technique.
Results
The evolution of the Pareto front during the optimisation
of the spherical tumour is shown in Figure 1. The
optimisation finished after 200 iterations (generations),
and so the final Pareto front represents the final results of
the optimiser.
Each point along the Pareto front represents a different
treatment plan. The front can be seen as a set of
compromises; it is impossible to decrease one objective
function without increasing another. This enables the user
to a posteriori decide relative objective importance and,
hence, choose the ideal treatment plan for each patient.
As an example, the treatment plan using 30 injections was
chosen. Its absorbed dose distribution through the central
slice is shown in Figure 2. To highlight the steep absorbed
dose gradient obtained with this treatment concentric
spherical shells, surrounding the tumour were also
included. Very satisfying treatment plans have been
defined using this new method of MBC.
Conclusion
A new form of BT, MBT, has been proposed, as well as a
promising method of generating optimal treatment plans.
It can be seen that the treatment plans proposed by the
optimiser (NSGA2) deliver satisfactory absorbed dose
distributions to the tumour, whilst sparing surrounding
tissue, which in turn spares more OARs. This method can
be used in real time during clinical treatment of MBT.
References
[1] K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan, “A
fast and elitist multiobjective genetic algorithm: NSGA-
II,” IEEE Trans. Evol. Comput. , vol. 6, no. 2, pp. 182–197,
2002.
PV-0184 Quantitative study on position margin in
Intraluminal Brachytherapy Planning for lung
treatment
C.W. Kong1, H. Geng1, Y.W. Ho1, W.W. Lam 1, K.Y.
Cheung1, S.K. Yu1
1Hong Kong Sanatorium & Hospital, Medical Physics and
Research Department, Happy Valley, Hong Kong SAR
China
Purpose or Objective
In Intraluminal Brachytherapy for lung treatment, a
Lumincath applicator, normally 5F flexible nylon catheter,
is inserted through the Trachea and Bronchus. High
activity radioactive source is loaded through the catheter
for treating the tumor site. Unlike external radiotherapy,
there is no motion control technique for afterloading
brachytherapy treatment. Breathing motion should affect
the position accuracy of Intraluminal Brachytherapy as
both Trachea and Bronchus move with the breathing
motion of the patient. It is not practical for the patient to
do breath-hold during treatment since the whole
treatment can last for several cycles of breathing
depending on the source activity. The additional margin
for treatment length should be considered in Intraluminal
Brachytherapy to compensate such effect. The objective
of this study is to investigate the position margin of
treatment planning on intraluminal brachytherapy for lung
treatment.
Material and Methods
We applied two-dimensional (2D) projection
reconstruction methods to measure the movement of
catheter due to the breathing motion. In 2D projection
reconstruction an orthogonal pair of isocentric
radiographs were taken on the patient inserted with the
Lumincath catheter. By localizing difference position
markers on the catheter in two separate projections, the
catheter can be reconstructed in three-dimensional (3D)
space for the planning calculation. The average position
difference of reconstructed points between two
projections reflects the accuracy of 2D reconstruction
method. By comparing the reconstruction accuracy
between two scenarios: patient doing free breathing and
breath-hold, the impact of breathing motion on the
position of catheter can be derived. In the study an
orthogonal pair of radiographs were done on patients with
free breathing and breath-hold; The discrepancy in the
average position difference between 2D projection
reconstructions with free breathing and breath-hold was