Abstract
BackgroundIntra-and inter-tumoral heterogeneity in growth dynamics and vascularity influence tumor response to radiation therapy. Quantitative imaging techniques capture these dynamics non-invasively, and these data can initialize and constrain predictive models of response on an individual basis.MethodsWe have developed a family of 10 biologically-based mathematical models describing the spatiotemporal dynamics of tumor volume fraction, blood volume fraction, and response to radiation therapy. To evaluate this family of models, rats (n = 13) with C6 gliomas were imaged with magnetic resonance imaging (MRI) three times before, and four times following a single fraction of 20 Gy or 40 Gy whole brain irradiation. The first five 3D time series data of tumor volume fraction, estimated from diffusion-weighted (DW-) MRI, and blood volume fraction, estimated from dynamic contrast-enhanced (DCE-) MRI, were used to calibrate tumor-specific model parameters. The most parsimonious and well calibrated of the 10 models, selected using the Akaike information criterion, was then utilized to predict future growth and response at the final two imaging time points. Model predictions were compared at the global level (percent error in tumor volume, and Dice coefficient) as well as at the local or voxel level (concordance correlation coefficient).ResultThe selected model resulted in < 12% error in tumor volume predictions, strong spatial agreement between predicted and observed tumor volumes (Dice coefficient > 0.74), and high level of agreement at the voxel level between the predicted and observed tumor volume fraction and blood volume fraction (concordance correlation coefficient > 0.77 and > 0.65, respectively).ConclusionsThis study demonstrates that serial quantitative MRI data collected before and following radiation therapy can be used to accurately predict tumor and vasculature response with a biologically-based mathematical model that is calibrated on an individual basis. To the best of our knowledge, this is the first effort to characterize the tumor and vasculature response to radiation therapy temporally and spatially using imaging-driven mathematical models.
Highlights
Intra-and inter-tumoral heterogeneity in growth dynamics and vascularity influence tumor response to radiation therapy
This study demonstrates that serial quantitative magnetic resonance imaging (MRI) data collected before and following radiation therapy can be used to accurately predict tumor and vasculature response with a biologically-based mathematical model that is calibrated on an individual basis
The quantitative and non-invasive nature of both diffusion weighted (DW-)MRI and DCEMRI makes these techniques well-suited for mathematical modeling of tumor growth as they can provide the 3D distribution of the tumor cells and vasculature before, during, and after treatment
Summary
Intra-and inter-tumoral heterogeneity in growth dynamics and vascularity influence tumor response to radiation therapy. Emerging imaging [4] and modeling approaches [5,6,7,8], may facilitate the improvement of radiation therapy through the assessment of intratumoral heterogeneity, the identification of tumor radiosensitivity, and through in silico trials to optimize therapeutic regimens (e.g., dosing and scheduling) for an individual subject Anatomical imaging approaches such as contrastenhanced magnetic resonance imaging (MRI), play a crucial role in identification of treatment volumes and the assessment of response [9]. Anatomical imaging does not assess physiological or functional properties of the tissue that might be relevant to tumor and radiation biology Quantitative imaging techniques such as diffusion weighted (DW-) MRI and dynamic contrast enhanced (DCE-) MRI can provide functional information characterizing changes in tumor cellularity and tissue perfusion, respectively [10]. These temporally defined data sets allow for model initialization and calibration, both of which are required for making tumor specific predictions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
