Inducing Biomechanical Heterogeneity in Brain Tumor Modeling by MR Elastography: Effects on Tumor Growth, Vascular Density and Delivery of Therapeutics

Heterogeneous Mechanical Stress and Interstitial Fluid Flow Predictions Derived from DCE-MRI for Rat U251N Orthotopic Gliomas

Mechanical stress and fluid flow influence glioma cell phenotype in vitro, but measuring these quantities in vivo continues to be challenging. The purpose of this study was to predict these quantities in vivo, thus providing insight into glioma physiology and potential mechanical biomarkers that may improve glioma detection, diagnosis, and treatment. Image-based finite element models of human U251N orthotopic glioma in athymic rats were developed to predict structural stress and interstitial flow in and around each animal's tumor. In addition to accounting for structural stress caused by tumor growth, our approach has the advantage of capturing fluid pressure-induced structural stress, which was informed by in vivo interstitial fluid pressure (IFP) measurements. Because gliomas and the brain are soft, elevated IFP contributed substantially to tumor structural stress, even inverting this stress from compressive to tensile in the most compliant cases. The combination of tumor growth and elevated IFP resulted in a concentration of structural stress near the tumor boundary where it has the greatest potential to influence cell proliferation and invasion. MRI-derived anatomical geometries and tissue property distributions resulted in heterogeneous interstitial fluid flow with local maxima near cerebrospinal fluid spaces, which may promote tumor invasion and hinder drug delivery. In addition, predicted structural stress and interstitial flow varied markedly between irradiated and radiation-naïve animals. Our modeling suggests that relative to tumors in stiffer tissues, gliomas experience unusual mechanical conditions with potentially important biological (e.g., proliferation and invasion) and clinical consequences (e.g., drug delivery and treatment monitoring).

Ultrasound Elastography Assessment of Knee Intra-Articular Adhesions at Varying Knee Angles

We aimed to verify the feasibility of using shear wave elastography (SWE) to quantify knee scars and the elastic modulus of scar tissues. Overall, 16 participants underwent SWE assessments and range-of-motion measurement and completed the Knee Injury and Osteoarthritis Outcome Score. The inter-rater reliability for SWE in the suprapatellar bursa, below the patellar tendon, and in the medial and lateral trochlear groove remained within 0.861-0.907. The SWE values in the four regions increased with increasing knee angle, and significant differences were observed between the values for below the patellar tendon and the suprapatellar bursa at knee flexion angles of 60° and 90°. The SWE values of the medial and lateral trochlear groove at 30°, 60°, and 90° knee flexion were higher on the affected side. A negative correlation was observed between the SWE values for the lateral trochlear groove at 0°, 30°, and 60° and those for below the patellar tendon at 0° and the suprapatellar bursa at 30° with both active and passive knee extension. The suprapatellar bursa value at 60° exhibited a positive correlation with both knee flexion and passive knee flexion, whereas that of the suprapatellar bursa at 90° exhibited a positive correlation with both the range of motion and passive range of motion. SWE is a replicable and effective method for detecting scar strength in the knee joint.

Investigating the synergistic effects of immunotherapy and normalization treatment in modulating tumor microenvironment and enhancing treatment efficacy

We developed a comprehensive mathematical model of cancer immunotherapy that takes into account: i) Immune checkpoint blockers (ICBs) and the interactions between cancer cells and the immune system, ii) characteristics of the tumor microenvironment, such as the tumor hydraulic conductivity, interstitial fluid pressure, and vascular permeability, iii) spatial and temporal variations of the modeled components within the tumor and the surrounding host tissue, iv) the transport of modeled components through the vasculature and between the tumor-host tissue with convection and diffusion, and v) modeling of the tumor draining lymph nodes were the antigen presentation and the development of cytotoxic immune cells take place. Our model successfully reproduced experimental data from various murine tumor types and predicted immune system profiling, which is challenging to achieve experimentally. It showed that combination of ICB therapy and normalization treatments, that aim to improve tumor perfusion, decreases interstitial fluid pressure and increases the concentration of both innate and adaptive immune cells at the tumor center rather than the periphery. Furthermore, using the model, we investigated the impact of modeled components on treatment outcomes. The analysis found that the number of functional vessels inside the tumor region and the ICB dose administered have the largest impact on treatment outcomes.

Tumor growth and population modeling in a toxicant-stressed random environment

When examining some factors that contribute to the growth or decline of a population or tumor, it is essential to consider a random hypothesis. By analyzing the effects of stress on a population (or volume of tumor growth) in a random environment, we develop stochastic models describing the dynamics of the population (or tumor growth) based on random adjustments to the population's intrinsic growth rate, carrying capacity, and harvesting efforts (or tumor treatments). Apart from the models' ability to capture fluctuations, the availability of a shape parameter in the models gives it the flexibility to describe a variety of population/tumor data with different shapes. The distribution of the stressed population size with or without harvesting (or treatments) is derived and used to calculate the maximum expected amount of harvests that can be taken from the population without depleting resources in the long run (or the minimum amount of chemotherapy needed to cause shrinkage or eradication of a tumor). The work done is applied to analyze tumor growth using published data comprising of the volume of breast tumor obtained by orthotopically implanting LM2-[Formula: see text] cells into the right inguinal mammary fat pads of 6- to 8-week-old female Severe Combined Immuno-Deficient mice.

Mathematical modeling of intratumoral immunotherapy yields strategies to improve the treatment outcomes

Intratumoral injection of immunotherapy aims to maximize its activity within the tumor. However, cytokines are cleared via tumor vessels and escape from the tumor periphery into the host-tissue, reducing efficacy and causing toxicity. Thus, understanding the determinants of the tumor and immune response to intratumoral immunotherapy should lead to better treatment outcomes. In this study, we developed a mechanistic mathematical model to determine the efficacy of intratumorally-injected conjugated-cytokines, accounting for properties of the tumor microenvironment and the conjugated-cytokines. The model explicitly incorporates i) the tumor vascular density and permeability and the tumor hydraulic conductivity, ii) conjugated-cytokines size and binding affinity as well as their clearance via the blood vessels and the surrounding tissue, and iii) immune cells-cancer cells interactions. Model simulations show how the properties of the tumor and of the conjugated-cytokines determine treatment outcomes and how selection of proper parameters can optimize therapy. A high tumor tissue hydraulic permeability allows for the uniform distribution of the cytokines into the tumor, whereas uniform tumor perfusion is required for sufficient access and activation of immune cells. The permeability of the tumor vessels affects the blood clearance of the cytokines and optimal values depend on the size of the conjugates. A size >5 nm in radius was found to be optimal, whereas the binding of conjugates should be high enough to prevent clearance from the tumor into the surrounding tissue. In conclusion, development of strategies to improve vessel perfusion and tissue hydraulic conductivity by reprogramming the microenvironment along with optimal design of conjugated-cytokines can enhance intratumoral immunotherapy.

MR Elastography in Cancer

ABSTRACT

The mechanical traits of cancer include abnormally high solid stress as well as drastic and spatially heterogeneous changes in intrinsic mechanical tissue properties. Whereas solid stress elicits mechanosensory signals promoting tumor progression, mechanical heterogeneity is conducive to cell unjamming and metastatic spread. This reductionist view of tumorigenesis and malignant transformation provides a generalized framework for understanding the physical principles of tumor aggressiveness and harnessing them as novel in vivo imaging markers. Magnetic resonance elastography is an emerging imaging technology for depicting the viscoelastic properties of biological soft tissues and clinically characterizing tumors in terms of their biomechanical properties. This review article presents recent technical developments, basic results, and clinical applications of magnetic resonance elastography in patients with malignant tumors.

Added Value of Viscoelasticity for MRI-Based Prediction of Ki-67 Expression of Hepatocellular Carcinoma Using a Deep Learning Combined Radiomics (DLCR) Model

Simple Summary

This study aimed to explore the added value of magnetic resonance elastography (MRE) in the prediction of Ki-67 expression in hepatocellular carcinoma (HCC) using a deep learning combined radiomics (DLCR) model. A total of 108 histopathology-proven HCC patients who underwent preoperative MRI and MR elastography were included. All the patients were divided into training and validation cohorts. An independent cohort including 43 patients was included for testing. A DLCR model was proposed to predict the expression of Ki-67 with conventional MRI (cMRI) as inputs. The images of shear wave speed (c-map) and phase angle (φ-map) derived from MRE were also fed into the DLCR model. Experimental results show that both c and φ values were ranked within the top six features for Ki-67 prediction with random forest selection, which revealed the value of MRE-based viscosity for the assessment of the tumor proliferation status in HCC. The model with all modalities (MRE, AFP, and cMRI) as inputs achieved the highest AUC of 0.90 ± 0.03 (CI: 0.89–0.91) in the validation cohort. The same finding was observed in the independent testing cohort with an AUC of 0.83 ± 0.03 (CI: 0.82–0.84). MRE-based c and φ-maps can serve as important parameters to assess the tumor proliferation status in HCC.

Abstract

This study aimed to explore the added value of viscoelasticity measured by magnetic resonance elastography (MRE) in the prediction of Ki-67 expression in hepatocellular carcinoma (HCC) using a deep learning combined radiomics (DLCR) model. This retrospective study included 108 histopathology-proven HCC patients (93 males; age, 59.6 ± 11.0 years) who underwent preoperative MRI and MR elastography. They were divided into training (n = 87; 61.0 ± 9.8 years) and testing (n = 21; 60.6 ± 10.1 years) cohorts. An independent validation cohort including 43 patients (60.1 ± 11.3 years) was included for testing. A DLCR model was proposed to predict the expression of Ki-67 with cMRI, including T2W, DW, and dynamic contrast enhancement (DCE) images as inputs. The images of the shear wave speed (c-map) and phase angle (φ-map) derived from MRE were also fed into the DLCR model. The Ki-67 expression was classified into low and high groups with a threshold of 20%. Both c and φ values were ranked within the top six features for Ki-67 prediction with random forest selection, which revealed the value of MRE-based viscosity for the assessment of tumor proliferation status in HCC. When comparing the six CNN models, Xception showed the best performance for classifying the Ki-67 expression, with an AUC of 0.80 ± 0.03 (CI: 0.79–0.81) and accuracy of 0.77 ± 0.04 (CI: 0.76–0.78) when cMRI were fed into the model. The model with all modalities (MRE, AFP, and cMRI) as inputs achieved the highest AUC of 0.90 ± 0.03 (CI: 0.89–0.91) in the validation cohort. The same finding was observed in the independent testing cohort, with an AUC of 0.83 ± 0.03 (CI: 0.82–0.84). The shear wave speed and phase angle improved the performance of the DLCR model significantly for Ki-67 prediction, suggesting that MRE-based c and φ-maps can serve as important parameters to assess the tumor proliferation status in HCC.