Speaker
Description
Spatiotemporal variations in hypoxia, cellularity, and vascularity affect the response of glioma to radiotherapy (RT). MRI can quantitatively map these features, identify tumor habitats, and calibrate biology-based models to predict the response to different RT protocols. Our long-term goal is to determine optimal RT strategies through predictive modeling.
Twenty-five female Wistar rats bearing C6 gliomas underwent multiparametric MRI every 72 hours for up to seven visits. RT was administered immediately after the first scan. Animals were assigned to the following groups: one 20 Gy dose (n=4), two 10 Gy doses (n=6), four 5 Gy doses (n=6), five 4 Gy doses (n=6), and a control group (n=4). For fractionated regimens, RT followed the corresponding imaging visit.
We employed five different approaches to analyze tumor habitats with ODE based models by varying the number of habitats (up to four) and number of parameters identified in the imaging data. All models were run on MATLAB (Mathworks, Natick, MA), using a finite difference approximation. The most parsimonious was selected via the Akaike Information Criterion (AIC).
The model with two habitats and two death terms when compared to the experimentally acquired data had the highest overall CCC value of 0.95 and the lowest AIC when averaged over the cohort. Future work will involve combining these approaches within a data assimilation framework to enable predictions of future habitat dynamics given previous tumor states.
Bibliography
@article{ostrom_cbtrus_2021,
title = {{CBTRUS} {Statistical} {Report}: {Primary} {Brain} and {Other} {Central} {Nervous} {System} {Tumors} {Diagnosed} in the {United} {States} in 2014–2018},
volume = {23},
copyright = {https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model},
issn = {1522-8517, 1523-5866},
shorttitle = {{CBTRUS} {Statistical} {Report}},
url = {https://academic.oup.com/neuro-oncology/article/23/Supplement_3/iii1/6381476},
doi = {10.1093/neuonc/noab200},
abstract = {Abstract
The Central Brain Tumor Registry of the United States (CBTRUS), in collaboration with the Centers for Disease Control and Prevention (CDC) and National Cancer Institute (NCI), is the largest population-based cancer registry focused exclusively on primary brain and other central nervous system (CNS) tumors in the United States (US) and represents the entire US population. This report contains the most up-to-date population-based data on primary brain tumors available and supersedes all previous reports in terms of completeness and accuracy and is the first CBTRUS Report to provide the distribution of molecular markers for selected brain and CNS tumor histologies. All rates are age-adjusted using the 2000 US standard population and presented per 100,000 population. The average annual age-adjusted incidence rate (AAAIR) of all malignant and non-malignant brain and other CNS tumors was 24.25 (Malignant AAAIR=7.06, Non-malignant AAAIR=17.18). This overall rate was higher in females compared to males (26.95 versus 21.35) and non-Hispanics compared to Hispanics (24.68 versus 22.12). The most commonly occurring malignant brain and other CNS tumor was glioblastoma (14.3\% of all tumors and 49.1\% of malignant tumors), and the most common non-malignant tumor was meningioma (39.0\% of all tumors and 54.5\% of non-malignant tumors). Glioblastoma was more common in males, and meningioma was more common in females. In children and adolescents (age 0–19 years), the incidence rate of all primary brain and other CNS tumors was 6.21. An estimated 88,190 new cases of malignant and non-malignant brain and other CNS tumors are expected to be diagnosed in the US population in 2021 (25,690 malignant and 62,500 non-malignant). There were 83,029 deaths attributed to malignant brain and other CNS tumors between 2014 and 2018. This represents an average annual mortality rate of 4.43 per 100,000 and an average of 16,606 deaths per year. The five-year relative survival rate following diagnosis of a malignant brain and other CNS tumor was 35.6\%, for a non-malignant brain and other CNS tumors the five-year relative survival rate was 91.8\%.},
language = {en},
number = {Supplement_3},
urldate = {2026-03-07},
journal = {Neuro-Oncology},
author = {Ostrom, Quinn T and Cioffi, Gino and Waite, Kristin and Kruchko, Carol and Barnholtz-Sloan, Jill S},
month = oct,
year = {2021},
pages = {iii1--iii105},
}
@article{marusyk_tumor_2010,
title = {Tumor heterogeneity: {Causes} and consequences},
volume = {1805},
copyright = {https://www.elsevier.com/tdm/userlicense/1.0/},
issn = {0304419X},
shorttitle = {Tumor heterogeneity},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0304419X09000742},
doi = {10.1016/j.bbcan.2009.11.002},
language = {en},
number = {1},
urldate = {2026-03-07},
journal = {Biochimica et Biophysica Acta (BBA) - Reviews on Cancer},
author = {Marusyk, Andriy and Polyak, Kornelia},
month = jan,
year = {2010},
pages = {105--117},
}
@article{syed_multiparametric_2020,
title = {Multiparametric {Analysis} of {Longitudinal} {Quantitative} {MRI} {Data} to {Identify} {Distinct} {Tumor} {Habitats} in {Preclinical} {Models} of {Breast} {Cancer}},
volume = {12},
issn = {2072-6694},
url = {https://www.mdpi.com/2072-6694/12/6/1682},
doi = {10.3390/cancers12061682},
abstract = {This study identifies physiological tumor habitats from quantitative magnetic resonance imaging (MRI) data and evaluates their alterations in response to therapy. Two models of breast cancer (BT-474 and MDA-MB-231) were imaged longitudinally with diffusion-weighted MRI and dynamic contrast-enhanced MRI to quantify tumor cellularity and vascularity, respectively, during treatment with trastuzumab or albumin-bound paclitaxel. Tumors were stained for anti-CD31, anti-Ki-67, and H\&E. Imaging and histology data were clustered to identify tumor habitats and percent tumor volume (MRI) or area (histology) of each habitat was quantified. Histological habitats were correlated with MRI habitats. Clustering of both the MRI and histology data yielded three clusters: high-vascularity high-cellularity (HV-HC), low-vascularity high-cellularity (LV-HC), and low-vascularity low-cellularity (LV-LC). At day 4, BT-474 tumors treated with trastuzumab showed a decrease in LV-HC (p = 0.03) and increase in HV-HC (p = 0.03) percent tumor volume compared to control. MDA-MB-231 tumors treated with low-dose albumin-bound paclitaxel showed a longitudinal decrease in LV-HC percent tumor volume at day 3 (p = 0.01). Positive correlations were found between histological and imaging-derived habitats: HV-HC (BT-474: p = 0.03), LV-HC (MDA-MB-231: p = 0.04), LV-LC (BT-474: p = 0.04; MDA-MB-231: p {\textless} 0.01). Physiologically distinct tumor habitats associated with therapeutic response were identified with MRI and histology data in preclinical models of breast cancer.},
language = {en},
number = {6},
urldate = {2026-03-07},
journal = {Cancers},
author = {Syed, Anum K. and Whisenant, Jennifer G. and Barnes, Stephanie L. and Sorace, Anna G. and Yankeelov, Thomas E.},
month = jun,
year = {2020},
pages = {1682},
}
@article{slavkova_mathematical_2023,
title = {Mathematical modelling of the dynamics of image-informed tumor habitats in a murine model of glioma},
volume = {13},
issn = {2045-2322},
url = {https://www.nature.com/articles/s41598-023-30010-6},
doi = {10.1038/s41598-023-30010-6},
abstract = {Abstract
Tumors exhibit high molecular, phenotypic, and physiological heterogeneity. In this effort, we employ quantitative magnetic resonance imaging (MRI) data to capture this heterogeneity through imaging-based subregions or “habitats” in a murine model of glioma. We then demonstrate the ability to model and predict the growth of the habitats using coupled ordinary differential equations (ODEs) in the presence and absence of radiotherapy. Female Wistar rats (N = 21) were inoculated intracranially with 10
6
C6 glioma cells, a subset of which received 20 Gy (N = 5) or 40 Gy (N = 8) of radiation. All rats underwent diffusion-weighted and dynamic contrast-enhanced MRI at up to seven time points. All MRI data at each visit were subsequently clustered using
k
-means to identify physiological tumor habitats. A family of four models consisting of three coupled ODEs were developed and calibrated to the habitat time series of control and treated rats and evaluated for predictive capability. The Akaike Information Criterion was used for model selection, and the normalized sum-of-square-error (SSE) was used to evaluate goodness-of-fit in model calibration and prediction. Three tumor habitats with significantly different imaging data characteristics (
p
{\textless} 0.05) were identified: high-vascularity high-cellularity, low-vascularity high-cellularity, and low-vascularity low-cellularity. Model selection resulted in a five-parameter model whose predictions of habitat dynamics yielded SSEs that were similar to the SSEs from the calibrated model. It is thus feasible to mathematically describe habitat dynamics in a preclinical model of glioma using biology-based ODEs, showing promise for forecasting heterogeneous tumor behavior.},
language = {en},
number = {1},
urldate = {2026-03-07},
journal = {Scientific Reports},
author = {Slavkova, Kalina P. and Patel, Sahil H. and Cacini, Zachary and Kazerouni, Anum S. and Gardner, Andrea L. and Yankeelov, Thomas E. and Hormuth, David A.},
month = feb,
year = {2023},
pages = {2916},
}
