Page 44 - 2014 Printable Abstract Book
P. 44
predisposed to poorer outcomes. The current standard of care for glioblastoma consists of
chemoradiation with standard doses and field margins in an attempt to capture invasive subclinical
disease, which ignores variability in tumor radiation sensitivity and growth kinetics, resulting in significant
volumes of normal brain tissue receiving large doses and limiting the dose delivered to the tumor. To this
end, we have focused our efforts to quantify the growth and response kinetics using routine clinical MRI
data and a patient-specific mathematical model. This patient-specific modeling approach quantifies
sensitivity to radiation therapy using the linear-quadratic radiobiological model and the patient’s own
treatment plan. The marriage of a patient-specific tumor model and the patient’s own clinical data
presents the potential of using a mathematical model and a multiobjective evolutionary algorithm (MOEA)
for intensity-modulated radiation therapy (IMRT) optimization to generate individualized radiation
treatment plans. By utilizing a patient-specific approach for radiotherapy plan generation on 9
glioblastoma patients, we find that the relative diffuse extent of each tumor quantified by the patient-
specific tumor model drives the non-uniform spatial distribution of optimized plans. The integrated model
takes advantage of patient-specific growth kinetics, metrics of response at multiple points in time and an
adaptive optimization process to generate plans that deliver dose to tumor much more efficiently than
the standard-of-care. With the increasing interest in high dose per fraction radiosurgery and heavy
charged particle therapies, we present patient-specific mathematical methods to analyze, predict and
interpret response to these radiation therapy modalities and their associated clinical trial designs.
(S604) Molecular imaging-based dose painting: challenges and opportunities for integration with
models of radiobiological response. Stephen Bowen, University of Washington, Seattle, WA
Dose painting represents the modulation of external beam radiation to account for differential
therapeutic response of cancers, which is driven in part by underlying spatial heterogeneity quantified on
cancer imaging. Cancer imaging modalities such as positron emission tomography (PET) can spatially map
in vivo distributions of metabolically active, proliferative, or hypoxic tumor regions, among many cancer
phenotypes, all of which have demonstrated prognostic value in risk stratification and predictive value in
therapeutic outcomes of individual patients. Quantitative PET for dose painting has emerged as a
powerful tool that is currently under investigation for efficacy in several prospective trials. Despite these
technical advances, several challenges must be addressed to make dose painting a clinical reality: (1) PET
visualization gap between cellular and millimeter tissue scales, (2) PET temporal gap between imaging
time points and underlying cancer growth/response dynamics, (3) transformation map of PET image
intensity to prescribed radiation dose magnitude for individual patients and tumor regions. This
presentation will review current dose painting techniques, including discrete tumor sub volume boosting
and continuous voxel dose redistribution. Opportunities for integration of dose painting with models of
radiobiological response will be discussed, focusing on variants of the linear-quadratic model. Lastly,
future directions in dose painting will be proposed that include both cancer and normal tissue imaging.
42 | P a g e
chemoradiation with standard doses and field margins in an attempt to capture invasive subclinical
disease, which ignores variability in tumor radiation sensitivity and growth kinetics, resulting in significant
volumes of normal brain tissue receiving large doses and limiting the dose delivered to the tumor. To this
end, we have focused our efforts to quantify the growth and response kinetics using routine clinical MRI
data and a patient-specific mathematical model. This patient-specific modeling approach quantifies
sensitivity to radiation therapy using the linear-quadratic radiobiological model and the patient’s own
treatment plan. The marriage of a patient-specific tumor model and the patient’s own clinical data
presents the potential of using a mathematical model and a multiobjective evolutionary algorithm (MOEA)
for intensity-modulated radiation therapy (IMRT) optimization to generate individualized radiation
treatment plans. By utilizing a patient-specific approach for radiotherapy plan generation on 9
glioblastoma patients, we find that the relative diffuse extent of each tumor quantified by the patient-
specific tumor model drives the non-uniform spatial distribution of optimized plans. The integrated model
takes advantage of patient-specific growth kinetics, metrics of response at multiple points in time and an
adaptive optimization process to generate plans that deliver dose to tumor much more efficiently than
the standard-of-care. With the increasing interest in high dose per fraction radiosurgery and heavy
charged particle therapies, we present patient-specific mathematical methods to analyze, predict and
interpret response to these radiation therapy modalities and their associated clinical trial designs.
(S604) Molecular imaging-based dose painting: challenges and opportunities for integration with
models of radiobiological response. Stephen Bowen, University of Washington, Seattle, WA
Dose painting represents the modulation of external beam radiation to account for differential
therapeutic response of cancers, which is driven in part by underlying spatial heterogeneity quantified on
cancer imaging. Cancer imaging modalities such as positron emission tomography (PET) can spatially map
in vivo distributions of metabolically active, proliferative, or hypoxic tumor regions, among many cancer
phenotypes, all of which have demonstrated prognostic value in risk stratification and predictive value in
therapeutic outcomes of individual patients. Quantitative PET for dose painting has emerged as a
powerful tool that is currently under investigation for efficacy in several prospective trials. Despite these
technical advances, several challenges must be addressed to make dose painting a clinical reality: (1) PET
visualization gap between cellular and millimeter tissue scales, (2) PET temporal gap between imaging
time points and underlying cancer growth/response dynamics, (3) transformation map of PET image
intensity to prescribed radiation dose magnitude for individual patients and tumor regions. This
presentation will review current dose painting techniques, including discrete tumor sub volume boosting
and continuous voxel dose redistribution. Opportunities for integration of dose painting with models of
radiobiological response will be discussed, focusing on variants of the linear-quadratic model. Lastly,
future directions in dose painting will be proposed that include both cancer and normal tissue imaging.
42 | P a g e
