82 / 1082
S69
ESTRO 36 2017
_______________________________________________________________________________________________
Conclusion
Incidental dose to the cardiac atria and ventricles did not
improve RP risk prediction in our cohort of s tage III NSCLC
patients as the DVH parameters for lung o utperformed
those for the heart. The multivariable mo del containing
the variables cardiac comorbidity and MLD is the optimal
model for RP prediction in this cohort.
OC-0143 Adaptive radiotherapy reduces pneumonitis
without increasing the risk of failure in lung cancer
A.A. Khalil
1
, M.M. Knap
1
, M.T. Petersen
1
, M. Kandi
1
, H.H.
Schmidt
1
, D.S. Møller
2
, L. Hoffman
2
1
Aarhus University Hospital, Department of Oncology,
Aarhus C, Denmark
2
Aarhus University Hospital, Department of Medical
Physics, Aarhus C, Denmark
THIS ABSTRACT FORMS PART OF THE MEDIA PROGRAMME
AND WILL BE AVAILABLE ON THE DAY OF ITS
PRESENTATION TO THE CONFERENCE.
OC-0144 Dosimetric analysis of randomized lung proton
and photon plans with respect to radiation toxicity
T. Deist
1
, P. Yang
2
, C. Oberije
1
, P. Allen
2
, Y. Luo
2
, Y.
Van Wijk
1
, D. Gomez
2
, T. X u
2
, S. Tucker
3
, R. Mohan
4
, S.
Hahn
2
, P. Lambin
1
, Z. Liao
2
1
MAASTRO Clinic, Department of Radiotherapy,
Maastricht, The Netherlands
2
The University of Texas MD Anderson Cancer Center,
Department of Radiation Oncology, Houston, USA
3
The University of Texas MD Anderson Cancer Center,
Department of Bioinformatics and Computational
Biology, Houston, USA
4
The University of Texas MD Anderson Cancer Center,
Department of Radiation Physics, Houston, USA
Purpose or Objective
Results from a Bayesian-randomized trial on intensity
modulated radiotherapy (IMRT) vs. passively scattered 3D
proton therapy (3DPT) show no significant difference in
protocol failure (i.e., either grade>=3 radiation
pneumonitis (RP) or local recurrence within 12 months).
We intend to analyze the differences in dose distribution
between modalities, relations between dosimetric
parameters and radiation-induced toxicities. The
objective is to identify dosimetric constraints that would
limit normal tissue complications in future trials.
Material and Methods
We analyzed 149 (57 3DPT, 92 IMRT) randomized trial
patients. DVH parameters for 3DPT and IMRT treatment
plans were compared for lung, esophagus, and heart. To
measure the predictive value of high- and low-dose
parameters for toxicity in lungs and esophagus, we fitted
V5 and V50Gy to RP and grade>=3 radiation esophagitis
(RE). Heart dose data was missing for 5 IMRT patients.
Results
RP and RE are not significantly different between
modalities (p>0.1, two-tailed Wilcoxon rank-sum test).
The difference between mean doses planned for IMRT and
3DPT plans was tested: Lung V5, V50, V60, V70Gy (in %),
esophageal V5Gy, and heart V5, V10Gy are significantly
different (p<0.005, two-tailed Wilcoxon rank-sum test).
The significant esophageal and heart dose parameters are
smaller for 3DPT, lung V5Gy is smaller, while lung
V50,V60,V70Gy are larger.
3DPT plan V5 and V50Gy are computed for each OAR and
compared to the respective median values of IMRT plans.
The percentage of 3DPT plans with V5Gy below or equal
to the IMRT median and V50Gy above the IMRT median are
reported:
<=V5Gy IMRT Median
>V50Gy IMRT Median
3DPT
Lung
0.93
0.68
Esophagus
0.67
0.49
Heart
0.93
0.54
3DPT plans yield smaller low dose regions in all three
OARs. However, 3DPT yields larger high dose regions in the
lungs.
To assess the relationship between low/high dose regions
and toxicity, V5 and V50Gy of lung and heart were fitted
to RP, esophageal V5 and V50Gy were fitted to RE:
There appears to be a stronger relationship between
toxicity and high dose regions in the affected OAR. Heart
doses have a weaker relationship with RP.
Conclusion
Tucker et al. (2013) showed that high dose regions in lung
tissue in lung IMRT have a pronounced effect on toxicity.
This is also observable in this trial cohort of IMRT and 3DPT
patients. In order to reduce toxicities, high dose regions
in normal tissues need to be reduced. 3DPT reduces low
dose regions significantly in all three OARs but high doses
regions are significantly higher in the lungs. Future
investigations should focus on stricter high dose
constraints for 3DPT plans. If such constraints are not
achievable due to penumbral and scattering
characteristics of protons and the usage of passive
scattering techniques, intensity modulated proton therapy
should be considered.
Joint Symposium: ESTRO-JASTRO: Oligometastatic
disease
SP-0145 Biological rationale and clinical evidence
P. Ost
1
1
University Hospital Ghent, Gent, Belgium
In 1995, Hellman and Weichselbaum hypoth esized that
the number and location of the metastases are an
important prognostic factor reflecting the metastatic
phenotype and hence prognosis of the cancer patient.
They suggested this based on the historical observation