453 / 1082
S438
ESTRO 36 2017
_______________________________________________________________________________________________
influence matrix). After solving the optimization
problems, the optimal fluence map is imported back to
Eclipse for leaf sequencing and final dose calculation using
the Eclipse API. The entire workflow is automated,
requiring user interaction solely to prepare the contours
and beam arrangement prior to launching the HCO Eclipse
API plugin. Optimization requires ~1-3 hours, after which
the automated plan including final dose calculation is
ready in Eclipse.
Results
HCO IMRT automatic planning was tested for 10 patients
with spinal lesions who had previously been treated to 24
Gy in a single fraction using either VMAT (8 patients) or
multi-field IMRT (2 patients). All automated HCO plans
used multi-field IMRT. A typical automated and clinical
plan comparison is shown in Figure 1, demonstrating
improved PTV coverage, cord and esophagus sparing with
the automated plan. As shown in Table 1, on average, the
automated plan improved PTV coverage (V95%) by 1%,
cord maximum dose by 2%, cord D0.35cc by 12%, cauda
maximum dose by 15%, and esophagus V18Gy by 100%. All
HCO plans met all clinical planning criteria.
Table-1.
Comparison of clinical and HCO automated plans
for ten patients. For each criterion, the better score is
bolded.
Figure-1.
Comparison of the clinical and automated plans
for a patient. A1-A3 represent the automated plan and C1-
C3 represent the clinical plan.
Conclusion
Hierarchical constrained optimization shows promise as a
powerful tool to automate IMRT treatment planning. The
automated treatment plan meets all clinical criteria and
compares favorably in relevant metrics to the plan
generated by planners. Using Eclipse API, we developed a
plugin which fully automates the workflow and can be
implemented into clinical use after thorough testing.
Poster: Physics track: Treatment planning: applications
PO-0824 IMRT dose painting for prostate cancer using
PSMA-PET/CT: a planning study based on histology
K. Koubar
1,2
, C. Zamboglou
2,3
, I. Sachpazidis
1,2
, R.
Wiehle
1,2
, S. Kirste
2,3
, V. Drendel
2,4
, M. Mix
2,5
, F.
Schiller
2,5
, P. Mavroidis
6,7
, P.T. Meyer
2,5
, A.L. Grosu
2,3
, D.
Baltas
1,2
1
Medical Center University of Freiburg - Faculty of
Medicine - University of Freiburg, Division of Medical
Physics - Department of Radiation Oncology, Freiburg,
Germany
2
German Cancer Consortium DKTK, Partner Site Freiburg,
Freiburg, Germany
3
Medical Center University of Freiburg - Faculty of
Medicine - University of Freiburg, Department of
Radiation Oncology, Freiburg, Germany
4
Medical Center University of Freiburg - Faculty of
Medicine - University of Freiburg, Department of
Pathology, Freiburg, Germany
5
Medical Center University of Freiburg - Faculty of
Medicine - University of Freiburg, Department of Nuclear
Medicine, Freiburg, Germany
6
University of North Carolina, Department of Radiation
Oncology, North Carolina, USA
7
Karolinska Institutet - Stockholm University,
Department of Medical Radiation Physics, Stockholm,
Sweden
Purpose or Objective
The goal of this work is to show the technical feasibility
and to evaluate the normal tissue complication probability
(NTCP) and the tumor control probability (TCP) of the
intensity modulated radiation therapy (IMRT) dose
painting technique using
68
Ga-HBED-CC PSMA-PET/CT in
patients with primary prostate cancer (PCa).
Material and Methods
We studied 10 RT plans of PCa patients having PSMA-
PET/CT scans prior to radical prostatectomy. One contour
was semi automatically generated for each patient on the
basis of the 30% of SUVmax within the prostate (GTV-PET).
For each patient, two IMRT plans were generated: PLAN
77
,
which consisted of whole-prostate radiation therapy to 77
Gy in 2.2 Gy per fraction; PLAN
95
, which consisted of
whole-prostate RT to 77 Gy in 2.2 Gy per fraction, and a
simultaneous integrated boost to the GTV-PET to 95 Gy in
2.71 Gy per fraction. The feasibility of these plans was
judged by their ability to reach prescription doses while
adhering to the FLAME trial protocol. Comparisons of TCPs
based on co-registered histology after prostatectomy
(TCP-histo) and normal tissue complication probabilities
(NTCP) for rectum and bladder were carried out between
the plans.
Results
Prescription doses were reached for all patients plans
while adhering to dose constraints. The mean doses on
GTV-histo for [Plan
77
and Plan
95
] were 75.8±0.3 Gy and
96.9±1 Gy, respectively. In addition, TCP-histo values for
Plan
77
and Plan
95
were 70±7 %, and 95.7±2 %, respectively.
PLAN
95
had significantly higher TCP-histo (p<0.0001)
values than PLAN
77
. There were no significant differences
in rectal (p=0.563) and bladder (p=0.3) NTCPs between
the 2 plans.
Conclusion
IMRT dose painting for primary PCa using
68
Ga-HBED-CC
PSMA-PET/CT was technically feasible. A dose escalation
on GTV-PET resulted in significantly higher TCPs without
higher NTCPs.
PO-0825 Multi-scenario sampling in robust proton
therapy treatment planning
E. Sterpin
1
, A. Barragan
2
, K. Souris
2
, J. Lee
2
1
KU Leuven, Department of Oncology, Leuven, Belgium
2
Université catholique de Louvain, Molecular imaging-
radiotherapy and oncology, Brussels, Belgium
Purpose or Objective
Beam specific PTVs (BSPTV) or robust optimizers are
superior to conventional PTVs for ensuring robustness of
proton therapy treatments. In these planning strategies,
realizations ('scenarios”) of a few types of uncertainties
are simulated: errors in patient setup, CT HU conversion
to stopping powers, and, more recently, breathing
motion. However, baseline shifts of mobile targets should
also be taken into account, which complicates the
sampling of the space of possible scenarios. We compare